Opportunities · Narrative Arcs
Macro narrative arcs under continuous tracking (event cluster → directional hypothesis → per-ETF breakdown). States escalate from nascent to actionable; decaying means the signal is fading. “Excess vs SPY” = the ETF's cumulative return over SPY since the narrative began: the higher, the more is already priced in.
Confidence range:0% – 100%192/192
Actionable 10
| Narrative arc | ETF | Direction | Confidence | Thesis | Excess vs SPY | First seen |
|---|---|---|---|---|---|---|
| Clean Energy Transition clean_energy_transition | GLD SPDR Gold Shares | Long | 0.75 | Maintain long. The prior thesis (2026-08-22, long 0.65) rested on three non-price quantitative links: (1) real fund creations/net inflows; (2) rising speculative net-long in CFTC; (3) a marginal easing in real rates lowering carry for non-yielding metals. Fresh independent data reinforce these links rather than invalidate them: CFTC net long is +141,648 contracts (report date 2026-08-18; four-period trend steadily higher; percentile=100), GLD AUM is $154.1B with net creations since 2026-08-10 of +$4,529M (+2.94% AUM) and 30-day inflows +3.45% AUM. 10y real rate = 2.35% (20d −8bp), easing the opportunity cost slightly. Price has already realized gains in this arc (5d excess +4.41%, window cum +4.41%, max DD −1.71%, last close 423.36), and option IV sits at a high percentile (96%) with technicals showing overbought conditions—signaling crowding and pullback risk. Net: non-price demand/positioning evidence supports staying long, but position sizing should be conservative and risk controls (stops, trimming rules) enforced because part of the move is already priced in. | +5.3% | 2026-08-17 |
| NATO / EU Defense Spending Cycle nato_eu_defense_spending | IAU iShares Gold Trust | Long | 0.70 | Maintain a medium-term (1–3 month) constructive stance on IAU. The prior view (small strengthen, confidence=0.65) argued geopolitical-driven safe-haven flows were underway and supported by independent non-price signals. Today's review keeps a long bias but slightly increases confidence to 0.70 because independent, non-price evidence strengthened: price_in excess (arc window) = +6.22%; ETF 8-day cum = +2.82%, max drawdown = -1.68%; CFTC COT net long = +141,648 contracts (wk +3,986 / +3%, at the 100th percentile) and OI = 406,260 (OI +21,657 vs 2026-07-28), indicating fresh, conviction-building positions; IAU net creations since 2026-08-14 = +$214M (+0.32% AUM) and 30d flow = -0.25% AUM (improved). 10Y real rate = 2.35% (20d -4bp) lowers the non‑carry cost of gold. These non-price signals independently support safe-haven demand even though much of the move is already price‑in (technicals show crowding: RSI14=70.9, distance to MA50 +10.57%, IV percentile 79%). Prediction-market odds of a NATO–Russia military clash remain low (~0.04), so escalation is not priced as an extreme tail. Net: independent supply/demand data strengthen the multi-week bullish thesis, but manage position sizing and hedges because of price run-up and crowding. | +6.4% | 2026-08-18 |
| Russia-Ukraine War russia_ukraine_war | BNO United States Brent Oil Fund LP | Long | 0.66 | Maintain a mid-term long on BNO. The prior assessment (2026-08-23, long, confidence 0.62) argued the same causal chain: conflict/sanctions → physical/near-month oil tightness → front-month backwardation → positive roll yield for short-dated oil ETF exposure. Re-evaluating adversarially, that chain remains valid and is reinforced by fresh non-price evidence. Key non-price signals: CFTC COT crude net-long +87,479 contracts (reporting week 2026-08-18; net positioning has held in the last four reports and rose +7,563 week-on-week, +10%), and the WTI term structure shows near-month > 12-month (85.3 vs 72.92 → annualized slope 16.978% as of 2026-08-20), consistent with physical tightness and positive roll dynamics. Price acts as corroboration: price_in 5d excess = +4.42% (BNO +3.52% / SPY -0.90%), arc window cumulative (2026-06-22→2026-08-18) BNO +20.85% vs SPY +3.10% (excess +17.75%), and etf_price_window (43d) cumulative +24.77%, max drawdown -15.21%. Historical analogs in this arc (27 events, ~81% bullish) align with the long view. Offsetting signals constrain conviction: OVX and option IV remain at subdued/normal levels (IV percentile 12%; OVX 49.63 z=-0.74), and short interest has materially risen (days-to-cover=1), increasing volatility and squeeze risk. Net effect: structural evidence strengthened the long thesis, warranting a modest confidence increase but with explicit risk triggers for reappraisal. | +21.9% | 2026-06-20 |
| China Financial Regulation china_financial_regulation | MCHI iShares MSCI China ETF | Long | 0.65 | Maintain medium-term (1–3 months) long. The published consultation on revising the Hang Seng Tech index (expanding 30→50, income-growth and hardware/AI tilt) reinforces the causal chain from index rule change → passive/rebalancing creation demand → re-rating for Hong Kong hard-tech and large China tech names. Non-price structural signals are mixed: options positioning is clearly call‑biased (put_call_vol_ratio ≈ 0.09, put_call_oi_ratio ≈ 0.29, call OI ≫ put OI; labeled call_crowded) and short interest is elevated (days_to_cover = 5.94; shorts +19.1% over 7 periods), implying crowded bullish positioning and asymmetric vulnerability to adverse headlines or delays. Fund flows are not confirming the demand story—MCHI AUM $6.3B with 30‑day net flows +0.01% AUM. Price has already partly priced the theme: arc window (2026‑06‑18→2026‑08‑18) MCHI +4.04% vs EEM −7.70% → MCHI‑EEM excess +11.74%; the recent 9‑day etf_price_window cumulative +0.07% with max drawdown −2.16% shows near‑term choppiness. Net: the consultation strengthens the structural thesis but is not final; without realized net creation and with option crowding, keep long at the same confidence (0.65) and monitor rule finalization, net creation/flows, and option OI dynamics before increasing exposure. | +2.9% | 2026-06-18 |
| US-Iran 2026 War us_iran_2026_war | EUAD Select STOXX Europe Aerospace & Defense ETF | Long | 0.63 | Maintain a medium-term long on EUAD, but with slightly reduced confidence. The WSJ reporting that an indefinite naval blockade of Iranian ports is an explicit policy option materially shortens the causal chain (conflict → shipping availability/insurance premium increase → higher defense/transport/energy valuation → EUAD benefit), which supports staying long. Price-derived evidence: arc window cumulative EUAD +18.59% vs ITA +12.13% (excess +6.47%); price_in excess = 10.2% (flagged=True), indicating material prior pricing; 9-day etf window cumulative = -2.14% with max drawdown -3.70% (creates a tactical entry opportunity). However, I must critique my prior assessment: the previous confidence lift overstated the near-zero probability of ground invasion—prediction-market data now shows manifold US-invade-before-end-2026 yes_prob = 0.12 (higher than prior), which weakens certainty that the conflict will remain limited to blockade actions. Non-price signals also weigh against increasing conviction: EUAD AUM experienced net outflows of -2.94% since 2026-08-10 (3 large creations/redemptions), and EUAD’s put-call skew is +1.86pt vs ITA -0.68pt (relatively higher demand for downside protection). Those items, together with price_in being flagged, mean part of the defensive/shipping-premium story is already priced and there is real money rotation. Therefore I keep the long stance but lower confidence to 0.63 and recommend holding with risk-managed sizing and staggered entries or hedges rather than expanding exposure immediately. | +14.2% | 2026-06-03 |
| NATO / EU Defense Spending Cycle nato_eu_defense_spending | GLD SPDR Gold Shares | Long | 0.55 | Maintain a mid-term (1–3 month) long on GLD. The arc transmission remains: geopolitical/defense stimulus → safe‑haven / allocation demand → ETF holdings rise. Non‑price evidence supporting this chain persists: arc price_in excess = +6.23% (GLD 8/18→8/21 vs absolute 0.0%), etf_price_window cumulative +2.81% with max drawdown −1.71% (partial price realization). Independently, CFTC COT net long remains elevated at +141,648 contracts (week +3,986; 100th percentile; OI=406,260) and GLD net creations since 2026‑08‑03 = +$2,015M (30d = +2.01% AUM), both mapping to allocation/hedge demand. Offsetting signals: option IV at the 96th percentile, RSI14=71, distance to MA50 +10.55% and a 1.65x volume surge — crowding / chase risk. Historical-analog strength is weak (this is a first_signal). Net: non‑price signals still favor long, price/technicals limit conviction → keep long at prior confidence 0.55. | +6.5% | 2026-08-18 |
| China Internet Platforms — Earnings/Users/Consumption china_internet_platforms | MCHI iShares MSCI China ETF | Long | 0.50 | Prior assessment (2026-08-20) strengthened a medium-term long on MCHI (confidence=0.55) on the structural thesis that the proposed Hang Seng Tech index expansion (30→50) would prompt passive/benchmark reallocation into Hong Kong tech names. That thesis relied on a non-price causal chain (index rule change → passive inflows → higher demand/liquidity for relevant constituents) plus modest real-money evidence (AUM ≈$6.2B, 30‑day flows +0.01% AUM) and favorable EM volatility regime. Critical re-check: the structural trigger remains present but has not produced a new, independent, quantifiable confirmation since the prior note (no announced implementation/dated inclusions). Current update keeps the long view because (1) non-price allocation metrics are not signaling outflows — AUM steady at $6.3B and 30‑day real flows +0.01% AUM (2026‑08‑20); (2) arc-level relative performance still favors MCHI: MCHI +5.48% vs EEM -5.18% → excess +10.66% since 2026‑06‑18; (3) EM volatility index VXEEM=27.05 (z=-2.06) remains in a low-vol regime supportive of risk-taking. Offsetting factors: short-term price action shows a 10‑day cumulative −2.23% and 10‑day max drawdown −4.41% (close 55.66), trading dollar volume is muted ($58.1M vs baseline $131.5M), narrative heat has accelerated (article count 2.2x over 5 days) and short-interest/days‑to‑cover remain elevated (15,353,069 shares; 5.94 days; recent +19.1%). These increase short-term volatility and crowding risk. Net: keep the medium-term long because the structural non-price thesis remains intact, but do not strengthen — no new independent non-price confirmation appeared; slightly lower conviction and maintain disciplined risk controls around earnings and macro windows. | +2.9% | 2026-06-18 |
| China Financial Regulation china_financial_regulation | FXI iShares China Large-Cap ETF | Long | 0.50 | Maintain a mid-term (1–3 month) long bias on FXI. The prior thesis correctly identified the mechanistic bullish path — a Hang Seng Tech index revision → passive/theme reallocation. The current consultation (proposal to expand constituents from 30 to 50 and add revenue-growth based inclusion) strengthens that structural channel: if implemented, concentration falls and more growth/hard-tech Hong Kong listings become eligible for passive/ETF inclusion, attracting incremental passive and thematic flows. However the consultation is not a fait accompli and passive flows are not yet visible (30d creation/redemption = +0.00% AUM; 90d = -1.09% AUM), while price has already materially outperformed (arc window FXI vs EEM excess +12.98%), so upside is partly priced. New non-price evidence nudges my view higher: a very low put_call_vol_ratio = 0.01 indicates unusually light downside hedging and a call-tilt in positioning, supportive for further upside in the short run but increasing crowding risk. Combining: (a) price-derived evidence — arc excess +12.98%; recent 9d etf window cum +0.56% with max drawdown -2.24%; (b) non-price evidence — put_call_vol_ratio 0.01, options IV percentile 29%, short interest 71,582,778 (+72.4% over 7 periods) days-to-cover 3.25; (c) lack of confirmed passive inflows — I recommend a conditional, disciplined long (add/hold) stance: structurally constructive, higher conviction than prior, but keep event-window hedges until rule implementation and actual net inflows are observed. | +5.2% | 2026-06-18 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | VGIT Vanguard Intermediate-Term Treasury ETF | Long | 0.42 | Maintain a mid-duration long on VGIT (1–3 months). The prior core logic—relying on non-price signals—remains intact: the market-implied front-end path is not priced materially hawkish (ZQ implied Fed funds path = 3.735%), Treasury volatility is low (MOVE = 73.4), and term premium/supply considerations continue to support the mid-end (10Y = 4.69%, term premium ≈ 0.8393%). Positioning is consistent: CFTC 5y COT net = -2,169,814 contracts with no extreme crowding (40th percentile), implying shorts exist but are not at an extreme that would preclude upside on a dovish surprise. Price cross-check supports relative resilience: arc-window VGIT vs TLT excess = +2.68% (VGIT -0.36% vs TLT -3.04%), 52-day cumulative -0.36%, max drawdown -1.59%. The today trigger is a company-level financing/talent poaching story (semiconductor startup raising ~$2bn) — per rules, financing news alone does not provide a credible, non-price causal channel to change Fed path or Treasury supply/demand absent additional, specific data. Therefore the event is neutral to the arc thesis and does not justify changing the prior view. Maintain long at confidence 0.42; will re-evaluate if US core PCE/GDP on 2026‑08‑26 prints materially differently than expectations. | -3.9% | 2026-06-07 |
| Metals Price & Demand metals_price_demand | GDXJ VanEck Junior Gold Miners ETF | Long | 0.35 | Maintain a medium-term long on GDXJ. The arc thesis—improving precious-metals demand and lower carrying costs driving miner/ETF appreciation—remains intact and is reinforced by new, non-price quantitative evidence. Price has materially re-priced the theme (price_in_excess ~33.57%: GDXJ 38.0% vs SPY 4.43%; etf_price_window cumulative +38.0% with max drawdown -11.79%; arc window GDXJ +25.29% vs SPY +5.38%, excess +19.91%), but secondary drivers support continuation: real, on-exchange ETF creations (net +$170M since 2026-08-14, +1.73% AUM; 30d +2.89% AUM), 10Y real rate ≈ 2.35% (≈ -8bp over 20d) which lowers the opportunity cost of gold, COT shows speculative net long (~137,662 contracts as of 8/13), and prediction markets price a high probability of August gold highs (polymarket yes_prob for $4,600 = 1.00; $4,700 = 0.81). These non-price signals form a plausible causal chain (demand/positioning + lower real rates → gold/miner flows → GDXJ). Key risks: high IV (80th pctile), RSI=72.2 and large prior price appreciation imply crowding/mean-reversion risk; position sizing should be staged and contingent on upcoming macro prints (e.g., 8/26 PCE/GDP data). | +33.6% | 2026-06-23 |
Confirming 36
| Narrative arc | ETF | Direction | Confidence | Thesis | Excess vs SPY | First seen |
|---|---|---|---|---|---|---|
| US-Iran 2026 War us_iran_2026_war | XLE Energy Select Sector SPDR Fund | Long | 0.82 | Maintain a medium-term long on XLE and modestly increase conviction. The prior view (2026-08-23, long 0.78) rested on a causal chain: US blockade → near-term transport/spot disruption → WTI near-term backwardation → rising speculative net-long. That causal chain still holds and is reinforced by today’s report (WSJ) that the US may keep a naval blockade of Iranian ports 'indefinitely', which directly supports sustained near-term supply friction. Structured, non-price evidence: WTI near/12m slope ≈ +17.02% (backwardation); CFTC crude net-long = +87,479 contracts (week +7,563; ≈ +10%), indicating speculative positioning building into tightness. Price cross-checks: XLE 9-day window cum = +4.45% (max drawdown -0.17%); arc cumulative XLE +7.22% vs SPY +6.89% (excess +0.34%) — the multi-day structure remains intact. Off-setting signals: prediction-market probability that a ceasefire continues to Aug 31 = 0.92 (makes a larger kinetic escalation less likely), XLE option IV percentile = 17% (market complacent), and 30-day ETF flows = -1.10% AUM (not a consistent buyer base). Conclusion: non-price, arc-relevant evidence (policy blockade signal + CFTC positioning) dominates and strengthens the prior thesis; raise confidence modestly from 0.78 to 0.82 while noting crowding/technical and high ceasefire-probability constraints. | +0.5% | 2026-05-14 |
| US-Iran 2026 War us_iran_2026_war | VDE Vanguard Energy ETF | Long | 0.82 | Maintain a medium-term long on VDE. Causal chain: the reported U.S. consideration of a prolonged naval blockade of Iranian ports maps directly to “shipping-lane disruption → near-term physical/delivery tightness → elevated oil risk premium → upstream/energy equities benefit.” Relevant non-price metrics already indicate tightness and are amplified by this event: WTI near-month vs 12-month curve in clear backwardation (slope ≈ +17.266%), CFTC speculative net-long ≈ +79,916 contracts, and EIA commercial crude stocks ≈ 723,104 kbbl — together these point to near-term physical tightness supportive of energy equities. Price/positioning corroborates (price_in excess 49d = +5.63%; etf_price_window 49d cumulative +9.21% with max drawdown −8.51%), showing market recognition of the risk premium. Short-term dampeners remain (prediction markets price high probability of ceasefire through 8/22–8/31, RSI14=71.5, ATM IV at low historical percentile 13%, short interest +40.5% with days-to-cover=1), which raise execution and volatility risk but do not negate the supply-disruption transmission mechanism. Therefore the blockade report strengthens the supply-disruption case and the VDE long, with active risk-management (staggered entries, explicit stops). | +5.6% | 2026-06-08 |
| US-Iran 2026 War us_iran_2026_war | GLD SPDR Gold Shares | Long | 0.80 | Maintain a medium-term long on GLD (1–3 months) and modestly strengthen conviction versus the prior note. Price evidence shows the market is repricing geopolitical risk (price_in 5d excess +5.31%; etf_price_window 9d cum +5.59% with max drawdown −1.71%; arc cumulative excess +1.22%), but per rules price is confirming, not deciding. Non-price quantitative support: CFTC COT net long rose to +141,648 contracts (report week 2026-08-18; at the 100th percentile; four-period trend +119,795→+130,766→+137,662→+141,648), indicating larger institutional/speculative net-long exposure; GLD institutional flows since 2026-08-10 show net creations +$4,529M (+2.94% AUM) and 30d net inflow +3.45% AUM, evidencing real funding demand; option-side ATM IV=27.26% (96th percentile) and a 1.65x intraday volume surge point to priced hedging demand. Offsetting signal: prediction markets assign high probability (yes_prob 0.92–1.00) that a ceasefire persists through late August, which tempers the tail-risk case for prolonged escalation. Real rates (10Y real = 2.35%, 20d −8bp) have moved in a direction supportive of gold. Net balance: structural, non-price positioning and flows favor a continued defensive long stance. Execute via phased entries and protection; re-evaluate around the 8/26 macro releases or if prediction markets materially re-price escalation probabilities. | -4.6% | 2026-06-03 |
| Latin America Economy latin_america_economy | TLT iShares 20+ Year Treasury Bond ETF | Long | 0.78 | Maintain a medium-term long on TLT. Causal chain: independent non-price signals point to real duration demand—since 2026-08-17 net creations/redemptions +$512M (+1.11% AUM) and 30‑day net inflows +9.29% AUM (as of 2026-08-21); CFTC 30y remains net-short but has been reducing shorts (latest net -361,383 contracts; weekly +3,441; at the recent percentile high), implying short covering and fresh long positioning; treasury auction metrics (bid-to-cover ~2.67) show genuine demand to absorb supply. These “to-the-point” non-price drivers form a transmission (flows → positioning/short-cover → duration bid) that supports TLT. Rate-market pricing is not decisively hawkish (implied Fed funds 3.74%, MOVE 73.4 lower, 10Y=4.69%, 2s10s=0.5), suggesting the move is more term-premium/supply-driven than a persistent policy shock. Price-derived signals show arc excess -0.97% (2026-07-29→2026-08-21) and 9‑day cum -0.17% with max drawdown -1.50%, so gains are not fully realized; therefore maintain long but cap position sizing/pace. Given the fresh independent non-price support, modestly increase conviction (+0.05). | -5.9% | 2026-07-29 |
| US-Iran 2026 War us_iran_2026_war | XOP SPDR S&P Oil & Gas Exploration & Production ETF | Long | 0.70 | Maintain a medium-term long on XOP (1–3 months). The causal chain still holds: US statements about a possible indefinite, rotating naval blockade of Iranian ports → potential reduction in near-term Iranian crude supply and shipping throughput → near-month crude tightness reflected in backwardation and rising speculative positioning → small/mid upstream E&P and oil services see magnified earnings/oil-price sensitivity, benefiting XOP. Non-price structural evidence supporting this chain includes CFTC crude net speculative longs at +87,479 contracts (report 2026-08-18; four‑period trend +79,916 → +87,479; 58th percentile) and WTI term structure showing ~17.02% annualized backwardation (near-month 85.33 vs 12‑month 72.92). However, I lower prior confidence because predictive markets price a high short-term probability of ceasefire/limited escalation (polymarket through Aug 31 yes_prob 0.92–1.00) and fund flows flipped recently: since 2026-08-14 XOP had five creation/redemption events net −$75M (−1.97% AUM) and 30‑day net −0.89% AUM, indicating recent withdrawal. Price-derived context: arc cumulative XOP +2.40% vs XLE +7.22% → XOP‑XLE excess −4.83%; 60‑day XOP cumulative +4.71% with max drawdown −15.46%; technicals show short-term overbought (RSI 71.5). Those price facts constrain sizing/timing but do not by themselves determine direction. Net: structural non-price drivers still lean positive → keep long, but reduce conviction and employ tranche entries and explicit add-on triggers tied to continuing COT/backwardation/EIA/OPEC signals. | -1.9% | 2026-06-06 |
| Metals Price & Demand metals_price_demand | GLD SPDR Gold Shares | Long | 0.68 | Maintain a medium-term (1–3 month) long stance on GLD. Causal chain: Iran attacks → safe‑haven demand → futures and ETF positioning builds → price and flows reflect demand. Non‑price quantitative evidence supports this: CFTC COT net long = +141,648 contracts, rising over the last 4 reports (week Δ +3,986, at the 100th percentile; OI = 406,260, +21,657 vs 2026‑07‑28, indicating new positioning), and GLD creations/net inflows since 2026‑08‑10 = +$4,529M (+2.94% AUM), 30d net inflow +3.45% AUM — real capital allocation into bullion. Price signals are consistent but secondary: 5‑day price_in excess = +5.31% (GLD +4.41% vs benchmark -0.90%), etf_price_window (9d) cumulative = +5.59% with max drawdown -1.71%. Macro context: 10y real rate = 2.35% (20d Δ -8bp) slightly reduces carry headwind. Prediction markets show extreme short‑term bullishness (polymarket $4,600 in August yes_prob = 1.00), reinforcing market expectations. Crowd risk is material: GLD options IV percentile = 96%, RSI14 = 71, single‑day USD volume 1.65x baseline, and crowded COT positioning all raise the probability of sharp mean reversion. Operationally: keep a long bias but manage sizing and use protective hedges/step‑in entries given elevated IV and positioning crowding. | +5.3% | 2026-06-14 |
| US-Iran 2026 War us_iran_2026_war | USO United States Oil Fund | Long | 0.66 | Maintain a medium-term long on USO. Causal chain: a U.S. naval blockade of Iranian ports held for an extended period (news: 'indefinitely' with ship rotations) increases shipping/insurance risk in Hormuz and nearby lanes → amplifies front-month physical tightness → yields pronounced WTI front‑vs‑12m backwardation, which benefits USO because it holds near‑month futures and realizes roll yield. Non-price quantitative support: CFTC crude net speculative long +87,479 contracts (report 2026-08-18; weekly +7,563/+10%; ~58th pctile) and a materially backwardated WTI curve (~+17.22% annualized front‑vs‑12m). Price cross-checks: arc window USO cumulative −11.48% (2026-05-04→2026-08-18) but a recent 9‑day etf_price_window cumulative +5.51% with max DD −2.02% (latest close 134.64 on 2026-08-21), indicating short‑term re‑pricing toward the news. Offsetting signals: prediction markets assign a high probability to ceasefire continuity (polymarket yes_prob=0.92 for through Aug 31) and USO has seen significant 30d/90d outflows (−14.36% / −29.68% AUM), which constrain upside and argue for conservative sizing. Net: today’s blockade report materially reinforces the prior causal thesis (so I modestly increase confidence to 0.66) but execution should be phased and monitored with EIA stocks, OPEC signals, and prediction‑market moves. | -15.4% | 2026-06-06 |
| BoJ Yen Normalization boj_yen_normalization | TLT iShares 20+ Year Treasury Bond ETF | Contested | 0.66 | Maintain 'contested' (two-sided) but reduce confidence. The prior core chain (BOJ/yen normalization → reduced Japanese selling pressure into U.S. Treasuries → lower supply risk → real money flows and auction absorption supporting duration) remains intact: independent non-price evidence is consistent with the chain — TLT AUM $46.9B (as of 2026-08-20), net real creations since 2026-08-14 +$824M (~+1.76% AUM), 30-day cumulative flows +9.24% AUM; recent Treasury auction bid-to-cover average ≈2.67 (strong demand); CFTC 30y net position -361,383 contracts (report week 2026-08-18), at the 100th percentile, with a small weekly net reduction of -3,441 contracts (+1%). These signals continue to support that demand can absorb supply. However, two weakening factors are material: (1) price has already reflected part of the thesis — price_in 5d excess = +1.65%, 8d window cum = +1.17%, 8d max drawdown = -1.50% — lowering remaining upside; (2) positioning is crowded — extreme COT net-short and rising FINRA short interest/days-to-cover imply both squeeze risk and elevated volatility. Net: non-price evidence remains supportive but not newly decisive; therefore keep contested and reduce confidence from 0.74 to 0.66. | -5.7% | 2026-06-14 |
| RBA Australia Rate Cycle rba_australia_macro_policy | QQQ Invesco QQQ Trust (Nasdaq-100) | Long | 0.65 | Maintain a mid-term long view on QQQ. The prior thesis argued macro_data_inflection combined with fund flows and extreme short positioning supported a tactical long. I revise one prior point: the earlier CFTC net-short extreme (-89,125 contracts as of 2026-08-11) was cited as a potential short-squeeze source; the latest COT shows net-short has narrowed to -61,771 contracts (week +27,354, +31%), indicating partial short covering and a changed positioning landscape — this weakens the ‘‘guaranteed squeeze’’ narrative but does not negate the bullish case. Non-price structural supports remain and are reinforced by the new event: (1) large net creations since 2026-08-03 of +$15,845M (+3.25% AUM, AUM $487.3B) show continued inflows; (2) option IV is low (IV historical pctile 16%; call IV 19.96%, put IV 18.0%), making downside protection cheap and lowering hedging costs; (3) volatility index VXN is depressed (23.26, z=-1.04), consistent with a calm risk environment favoring risk assets. Price action shows only a mild pullback (etf_price_window 9d cum -1.38%, max drawdown -2.89%, price_in excess (3d) +0.07%). The triggering macro_data_inflection event (UK Q2 growth example) reinforces global risk-on expectations that prop up growth/tech multiple expansion. Event risks (upcoming NVDA/AVGO earnings, AU CPI on 2026-08-26 potentially affecting RBA path) remain reasons to manage sizing; overall confidence is moderately high (0.65). | +0.1% | 2026-08-19 |
| US-Iran 2026 War us_iran_2026_war | BNO United States Brent Oil Fund LP | Long | 0.65 | Maintain a medium-term long on BNO. The latest US statements that a naval blockade of Iranian ports could be “indefinite / rotational” strengthen the causal chain: shipping-lane disruption → near-term physical/arrival tightness → front-month backwardation → positive roll/risk premium for Brent-linked/near-month-exposed ETFs like BNO. Non-price quantitative backing: CFTC speculative net long remains elevated and has not withdrawn over the last four reports (+92,943 → +86,958 → +79,916 → +87,479 contracts, report dated 2026-08-18), and the WTI front-vs-12m curve shows significant backwardation (~17.27% annualized), which is the direct mechanism by which shipping disruption converts to ETF roll gains. Price evidence is supportive but secondary: recent 5-day BNO excess ≈ +2.66% (2026-08-18), 52-day etf_window cumulative +3.68% with a historical max drawdown of -24.05%; arc window since 2026-06-08 shows BNO +0.42% vs SPY +3.82% (excess -3.39%), indicating relative underperformance. Offsetting constraints: prediction markets assign high probability to near-term ceasefire/containment (polymarket 'ceasefire through Aug31' yes_prob ≈ 0.92), BNO ATM IV is low (IV percentile ~12%), and short interest has risen sharply (+48.4%, days-to-cover=1), implying crowding and squeeze/volatility risk. Net: the blockade rhetoric strengthens the structural long thesis, so confidence is raised from the prior 0.58 to 0.65, but disciplined event-based risk controls remain necessary because market pricing is still complacent on major escalation. | +0.1% | 2026-06-08 |
| Korea / US Geopolitics korea_us_geopolitics | IAU iShares Gold Trust | Long | 0.60 | Maintain a medium-term (1–3 month) long on IAU but slightly lower conviction. Non-price, “on‑point” evidence still supports bullish exposure: CFTC gold net long = +141,648 contracts (report 2026-08-18; four-period trend rising, weekly +3%), 10Y real rate = 2.35% (20d Δ = -8bp) which modestly eases the carry penalty for non‑yielding metals, and fund flows: IAU AUM $68.1B with net creations +$315M (+0.46% AUM) since 2026-08-17. Price-derived metrics also show realized strength: price_in excess = +6.65% (52‑day window cumulative), etf_window cumulative +6.65%, max drawdown -8.22%. These signals collectively point to an active safe‑haven/positioning bid, so I keep the long. However, the triggering item today (event_opp #4166 — a Carson Block invited piece about passive investing and market structure) is not a Korea–US geopolitical escalation and therefore does not strengthen the arc’s causal chain. By contrast, prior geopolitics-aligned events in the arc have coincided with notable IAU upticks (e.g., 2026-08-15 event windows showing +5.6% / +7.3% on 9-day windows), demonstrating that true geopolitical shocks lift gold; today’s article does not provide that channel. Balancing the “on‑point” non-price evidence against the fact that the new trigger is off‑arc and given elevated option IV (ATM IV 28.9%, IV percentile 79%) and technical/positioning crowding (RSI14=70.9, price +10.6% vs MA50), I reduce conviction from 0.65 to 0.60. Continue to hold/size modestly long; reduce or hedge if COT retracts sharply, 10Y real rate spikes materially, or risk‑on sentiment reverses. | +3.1% | 2026-08-17 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | KRE SPDR S&P Regional Banking ETF | Short | 0.60 | Maintain a medium-term (1–3 month) short on KRE, with stronger conviction than prior. Causal chain: (1) Fund flows — since 2026-08-14 KRE has seen real redemptions of -$427M (~-10.51% AUM), 30-day cumulative outflow -8.69% AUM, a direct non-price quantitative signal indicating investors are exiting regional-bank exposure and creating selling pressure on the ETF; (2) Macro expectations — prediction market probability of another 2026 Fed hike ~0.47 (manifold) does not materially raise the odds of a higher-for-longer path, supporting the thesis that a more dovish/softer rate path compresses bank NIM and weighs on regional-bank valuations; (3) Price context and risk — price_in excess = -4.21% (KRE +6.4% vs XLF +10.6%), etf_window cumulative +6.4% and max drawdown -4.13% show partial downward repricing but room remains. Offsetting factors: WALCL 4-week -$125B could transiently lift short/mid rates (positive for bank earnings), short interest modestly down (-2.5%) and low IV/negative skew reduce immediacy of a panic move. Net assessment: the large, targeted fund outflows are the dominant non-price evidence and strengthen the short thesis. The triggering item (a semiconductor startup raising capital) is not causally connected to the Fed→NIM→regional-bank chain and is therefore neutral/noise for this arc. | +2.8% | 2026-06-07 |
| Metals Price & Demand metals_price_demand | IAU iShares Gold Trust | Long | 0.60 | Conclusion: With multiple non-price quantitative signals converging toward the bullish side, the arc's short-to-medium-term tilt for IAU moves from contested to biased long. Rationale: 1) Price context shows a meaningful realized move — over the latest 47-day window IAU has an excess return of +6.81% (window cumulative +6.81%, max drawdown -8.22%), indicating an established up-leg but with retracement risk; 2) Non-price evidence has shifted toward longs: the latest CFTC COT (2026-08-18) reports net long +141,648 contracts, rising across four reports (weekly +3,986, at the sample's 100th percentile) and total OI increased to 406,260 (+21,657 vs 2026-07-28) — consistent with new conviction buying; 3) Real yields have eased in the short run (10y real yield = 2.35%, 20d -8bp vs the prior report's short-term uptick), lowering the opportunity cost of holding gold; 4) ETF demand signal: IAU AUM $68.1B and net creations since 2026-08-17 of +$315M (+0.46% AUM) show recent investor inflows beyond mere mark-to-market; 5) Prediction markets show high short-term bullish probabilities (Polymarket yes_prob 0.81–1.00 for August highs), adding to speculative demand evidence. Risks: option IV percentile is high (79%), RSI=70.9 and price distance to MA50 = +10.6% indicate crowding/overbought risk; upcoming core PCE and other macro prints could re-elevate real yields and quickly reverse positioning. Decision is grounded in converging non-price signals (COT, real yields, ETF creations) while using price history (excess +6.81%, drawdown -8.22%) as context; therefore we strengthen toward long but keep moderate confidence given crowding and near-term macro event risk. | +5.4% | 2026-06-14 |
| BoJ Yen Normalization boj_yen_normalization | VNQ Vanguard Real Estate ETF | Long | 0.60 | Maintain a long stance on VNQ (duration-sensitive/relative long) but with slightly reduced conviction. The prior structural chain (U.S.–Japan FX interventions and dollar weakening reduce the immediate risk of forced Japanese selling of foreign assets → lower likelihood of a sharp rise in U.S. yields → supports duration-sensitive assets like VNQ) remains intact (event_opp #4087). Non-price quantitative corroboration still exists: Fed net liquidity (WALCL−TGA−RRP) is $5,792B as of 2026-08-19, with the 4-week withdrawal moderating to -$125B; and VNQ short interest fell 9.0% to 5,506,617 shares (days-to-cover 1.9), reducing crowded-short dynamics. Price-derived facts show the arc window cumulative: VNQ +0.59% vs SPY +2.39% → price_in excess = -1.80% and max drawdown = -4.53%; this is an improvement versus the prior price_in (≈ -3.8%), meaning part of the thesis has been priced in and conviction should be trimmed. Options IV sits at low percentiles (ATM IV low, IV pctile 8%), which is consistent with a calmer risk backdrop but cannot alone determine direction. Conclusion: structural rationale still points to long, but partial price realization and time decay warrant weakening conviction. | -2.3% | 2026-07-02 |
| BoJ Yen Normalization boj_yen_normalization | VGIT Vanguard Intermediate-Term Treasury ETF | Long | 0.55 | Maintain a medium-term bias long VGIT, but with reduced conviction. The prior thesis argued BOJ/JPY normalization and related interventions reduce the tail risk of Japan-driven forced selling of U.S. Treasuries, thereby lessening systematic selling pressure on intermediate-duration (VGIT) vs long-duration (TLT). That thesis rested on non-price signals (CFTC 5Y positioning unwind, strong auction demand, low MOVE) and price had already partially reflected the view. Today the structural chain remains intact (JPY stability after intervention; auction demand remains strong; MOVE low), so the directional call stays long. However, conviction is reduced for three reasons: 1) a key non-price indicator (CFTC 5Y) no longer shows further net-short unwind — net position is -2,169,814 contracts, w/w -22,070, pctile=40, weakening the prior 'short-covering' narrative; 2) price_in has already realized part of the expected excess (VGIT vs TLT excess +3.33%; VGIT window cumulative -0.95%; max drawdown -1.59%), leaving less upside; 3) technicals and short-term flows do not provide additional multi-directional confirmation (RSI14=44.8; MA20<MA50; daily volume $289M vs 20d $188.4M = 1.54x, neutral). Net: structural, non-price support (e.g., strong auction bid-to-cover ≈2.67; MOVE=73.4) still underpins the long call, but because COT and realized price moves weaken the marginal case, confidence is lowered from 0.65 to 0.55 and we avoid increasing exposure ahead of upcoming macro prints. | -2.4% | 2026-06-14 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | QQQ Invesco QQQ Trust (Nasdaq-100) | Long | 0.55 | I explicitly acknowledge the prior decision (2026-08-22 long, confidence 0.50): it rested on non-price signals (prediction market showing low odds of further Fed hikes, system liquidity high) plus flows and positioning (net ETF creation and COT structure). Critique of the prior: the prior relied materially on recent flows/positioning which could be noisy; MarketWatch-style qualitative pieces were correctly treated as non-decisive. Current assessment: the independent non-price evidence has persisted and modestly strengthened, so I keep a mid-term long on QQQ and raise conviction slightly. Quantitative grounding: price_in excess = -1.13% (QQQ +4.67% vs SPY +5.80%) — prices have not structurally invalidated the thesis; etf_window cumulative +4.67% with max drawdown -11.32%. Independent non-price signals supportive of long: QQQ net creation since 2026-08-10 = +$9,788M (+2.01% AUM), 30-day net inflows +2.74% AUM; CFTC (nasdaq) net position = -61,771 contracts but weekly change +27,354 (+31%), at the 89th percentile (crowded shorts with recent short-covering); ATM option IV at 16th historical percentile (cheap protection). Absent a new, on-the-record hawkish Fed surprise or other countervailing non-price evidence (e.g., large equity financing that reliably increases supply), the available independent signals justify a small increase in conviction (0.55). Key near-term risks: US core PCE/GDP in ~2 days — a hawkish surprise would be a credible flip trigger. | -1.1% | 2026-05-14 |
| Russia-Ukraine War russia_ukraine_war | XLE Energy Select Sector SPDR Fund | Long | 0.55 | Maintain a medium-term long on XLE (1–3 months). Prior recommendation rested on two non-price pillars: CFTC speculative net longs and WTI futures backwardation indicating near-term physical tightness/positive roll. Re-check: (a) the historical analog (COT net-long + backwardated curve → tighter physical/roll-support for oil-sensitive equities) still holds; (b) price action corroborates the bullish side within this arc — XLE arc excess ≈ +7.11% (XLE +10.50% vs SPY +3.39%) and the 9-day etf_price_window shows +4.45% cumulative with max drawdown -0.17%; (c) the new event (conflict_escalation: Zelensky’s 40-day drone surge) reinforces the causal chain (escalation → tighter short-term physical/transport / higher risk premium → continued backwardation and speculative buying → supportive for energy producers). The primary non-price evidence is: CFTC net speculative long = 87,479 contracts (report 2026-08-18, +10% WoW) and a materially backwardated WTI curve (near-month 85.3 vs 12-month 72.92; 12m slope ≈ 16.98%). These form the basis for a strengthened long view. Offsetting signals (low options IV percentile = 17%, RSI 71.7, 30d ETF net flow -1.10% AUM) raise crowding/valuation caution, so I increase confidence modestly to 0.55 rather than a large incremental bet. | +7.3% | 2026-07-20 |
| BoJ Yen Normalization boj_yen_normalization | VTI Vanguard Total Stock Market ETF | Long | 0.55 | Maintain a mild medium-term (1–3 month) long on VTI. Causal chain: BOJ/US-Japan intervention → JPY stabilization → reduced forced Japanese selling of US assets for FX needs → less exogenous selling pressure on US equities, supporting broad US equity ETFs. Non-price quantitative support: CFTC (S&P) shows short-covering signs (net short = -280,446 contracts; weekly change +49,553, near recent highs), indicating short reduction/risk-on; market volatility is low (VIX=14.89, z=-1.20) and option IV sits at a mid-percentile (IV_pct=46%) with a modest 5-day IV uptick (+1.06pt), so risk premia have not materially widened; Fed net liquidity remains large but is contracting on the margin (WALCL−TGA−RRP = $5,792B; 4-week Δ = -$125B). Price-derived cross-checks show limited realization: arc window excess return = +0.06% (VTI +1.09% vs SPY +1.03%); etf_window cumulative +1.09%, max drawdown -3.29%, and recent price moved 379.04 → 376.58 (-0.64%). Structural non-price signals remain mildly supportive but not freshly decisive, so keep a gentle long posture while monitoring upcoming macro events (core PCE, CPI/NFP) and liquidity trends. | +0.1% | 2026-06-14 |
| BoJ Yen Normalization boj_yen_normalization | VOO Vanguard S&P 500 ETF | Long | 0.50 | Maintain a medium-term (1–3 months) long view on VOO. Transmission: BOJ/FX intervention that stabilizes the yen can reduce global risk premium → risk appetite returns → US equities supported. Price context: arc excess return is +1.03% with a max drawdown of -3.4% (price_in/etf_window provide context but are not the direction source). Non-price, quantitative support: CFTC S&P futures net short has retraced to -280,446 contracts (w/w +49,553, at the 100th percentile), indicating short-covering; Fed balance-sheet related liquidity (WALCL 4-week) has improved relative to the last report to -125B (total $5,792B as of 2026-08-19), lessening liquidity withdrawal risks. Constraints: option IV sits at a low historical percentile (IV pctile 17%), option skew lacks proven directional mapping, and a 1.54x volume surge with price down signals short-term distribution. Historical analogs remain first_signal/unclear. Conclusion: independent non-price data (COT + WALCL improvement) reinforce the prior BOJ/FX → risk-on transmission, so modestly strengthen the long stance (confidence 0.50 vs prior 0.45) while keeping position sizing cautious given the partial price giveback and weak analog strength. | -0.0% | 2026-06-14 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | VTI Vanguard Total Stock Market ETF | Long | 0.48 | Maintain a medium-term long on VTI. Prior assessment (2026-08-23) was also long with confidence 0.48 based on: (1) a highly net-short S&P futures position (CFTC net = -281,402 contracts, 98th percentile) implying potential short-covering support; (2) systemic liquidity contraction (WALCL−TGA−RRP = $5,792B, 4-week change -$125B) and a ~0.47 market-implied probability of further 2026 hikes—these non-price quantitative signals limit but do not reverse a long bias. Today's trigger (a chip startup poaching Nvidia talent) reinforces the ai_compute_capex_race narrative but provides no new non-price quantitative evidence that would alter the Fed-rate-path thesis. Price cross-check: price_in_excess = +0.19% (VTI +3.78% vs SPY +3.58%), window cum = +3.78%, max drawdown = -3.29%; options IV is at a low historical percentile (IV pctile = 0%) and put-call skew = -2.52pt (relatively more bullish vs SPY). Conclusion: non-price metrics still favor holding the long (short-cover risk balanced vs liquidity tightening); the startup story is a thematic reinforcement, not a macro trigger—so keep the prior stance and confidence. | +0.2% | 2026-06-07 |
| Metals Price & Demand metals_price_demand | SIL Global X Silver Miners ETF | Long | 0.42 | Remain long (moderate-to-strong conviction but not high). Causal chain: August’s renewed gold strength (news: 'gold shining again') transmits to miner/bullion equity ETFs like SIL because SIL is tightly correlated with gold (vs GLD = 0.91); higher gold demand/pricing should lift miner exposures. Non-price quantitative supports: (1) FRED 10Y real yield = 2.35%, 20d change -8bp (lower real yields reduce the opportunity cost of non-yielding metals, a direct macro driver); (2) real fund flows into SIL: +$77M since 2026-08-10 (+1.45% AUM) and 30d cumulative +0.89% AUM, indicating continued institutional buying; (3) prediction markets showing extreme short-term gold bullishness (polymarket: $4,600 in August yes_prob = 1.00), an independent probability signal supporting near-term gold upside. Price evidence corroborates (price_in excess = +13.05%; etf_window cumulative = +14.49%; max drawdown = -19.97%; latest close 99.63) but per rules price is supportive context only. Risks: option IV at 96th percentile and RSI14 = 72 indicate crowding/overbought; therefore position sizing should account for elevated tail risk and monitor real yields, flows, and prediction-market signals for confirmation or reversal. | +13.1% | 2026-08-14 |
| Russia-Ukraine War russia_ukraine_war | VDE Vanguard Energy ETF | Long | 0.40 | Maintain a mid-term long bias on VDE but reduce conviction and recommend smaller, staggered sizing with tighter stops. Rationale (quantified + transmission): 1) Non-price signals still support tighter energy fundamentals — WTI futures curve in clear backwardation (front vs 12m, annualized slope ≈16.978%) and recent CFTC speculative net-long positions around ~80k–87k contracts (e.g., ~86,958), implying tighter physical oil and positive roll yield which transmits to energy-company margins and VDE upside; 2) Those non-price signals have largely been priced in: over the 40-day arc window VDE cumulative +16.8% with an excess vs SPY of +13.93% and max drawdown only -3.96%; 3) Risk/crowding indicators have risen — RSI14=71.5 (overbought), distance to MA50 +10.02%, low ATM IV percentile (13%) with call-side crowding, and short interest up +40.5% (259,228 shares) with days-to-cover=1 indicating squeeze/liquidity risk; 4) The trigger event (#4075) is propaganda/conflict rhetoric and supplies no quantifiable supply-disruption evidence, so it neither strengthens nor contradicts the core non-price signals. Net: structural thesis still favors energy exposure, but with reduced conviction because much of the upside is realized and crowding/liquidity risks are elevated. | +13.9% | 2026-06-20 |
| BoJ Yen Normalization boj_yen_normalization | XLRE Real Estate Select Sector SPDR Fund | Long | 0.40 | Maintain a medium-term bullish view on XLRE driven by the arc thesis that BOJ/FX intervention reduces the likelihood of yen-driven rapid normalization of rates and thus limits downside for rate-sensitive sectors. The central policy signal (central_bank_intervention_fx) that underpins this causal chain remains intact. However, since the prior update (2026-08-18) independent non-price metrics weaken conviction: 1) fund flows show net creations/redemptions of -$155M since 2026-08-12 (-1.82% AUM) with 30d -0.71% AUM and 90d -0.76% AUM indicating capital is leaving the ETF; 2) short interest has been accumulating (7,454,300 shares, +4.6% vs prior; 7-period cumulative +16.4%; days-to-cover 1.56), raising downside pressure and crowding risk; 3) options IV is very low (IV percentile 8%), meaning downside protection is cheap and the market is complacent; 4) price has partly realized the thesis (price_in excess = -1.50%; etf_window cumulative +0.90%, max drawdown -4.19%), reducing remaining edge. Net: policy signal still supports long, but flow and positioning evidence argue for reduced conviction (0.40) and tighter risk controls. | -1.9% | 2026-07-02 |
| BoJ Yen Normalization boj_yen_normalization | KBE SPDR S&P Bank ETF | Flat | 0.40 | Remain flat. The prior assessment argued that there is no arc-specific, quantifiable non-price evidence that connects a BOJ/JPY normalization to KBE (e.g., sustained JGB yield rise, persistent JPY COT net-long, or an explicit BOJ policy shift). Since the last update, KBE has continued to underperform XLF (arc-window excess = -3.31%; recent 8-day cumulative = -1.17%, max drawdown = -3.56%), but per rules price alone is not dispositive. Non-price snapshot is mixed and does not give a clear directional signal: net creations/redemptions since 2026-08-13 = -$14M (-0.81% AUM) while 30d cumulative flows = +6.44% AUM; option IV is elevated (ATM IV percentile 86%; call IV 36.57% vs put IV 25.37%, skew -11.21pt) but chain quality is sparse; short interest = 15,145,952 shares (near-7-period change +265.2%, days-to-cover = 11.2) implies crowded shorts/squeeze risk; Fed net liquidity down -$125B (WALCL−TGA−RRP = $5,792B) exerts mild headwind for risk. Historical analogs inside the arc lean bearish (42 events: 11 long / 29 short / 2 mixed). No new, arc-specific non-price evidence emerged to justify taking a position, so remain flat and reduce conviction from 0.45 to 0.40 given elevated IV, crowded shorts and liquidity reduction. | +2.6% | 2026-06-14 |
| Latin America Economy latin_america_economy | QQQ Invesco QQQ Trust (Nasdaq-100) | Long | 0.36 | The prior assessment was contested (confidence 0.28) because funding/DRAM spot supported a small tech long but COT and growing short interest provided strong counterevidence. In this refresh the independent non-price signals tilt toward a modestly bullish stance: QQQ has seen net creation inflows of +$9,788M since 2026-08-10 (+2.01% AUM) and 30-day net flows +2.74% AUM, indicating active capital buying rather than a pure price derivative; CFTC COT (report 2026-08-18) remains net-short but shows a week-on-week reduction in net-short exposure (net -61,771 contracts, week +27,354), consistent with partial short-covering and reduced extreme crowding risk; DRAM spot (main series +9.16%) and low ATM option IV (19.96%, 16th percentile) and a subdued VXN (23.26, z=-1.04) together support a demand/risk-on backdrop for technology/semiconductor exposure. Price has already partly realized the view: arc window (2026-07-29→2026-08-21) QQQ +7.81% vs SPY +4.97% → price_in excess = +2.84%; the recent 9-day etf_window is -0.7% cumulative with -2.89% drawdown, showing ongoing short-term volatility. Therefore I move to a modest long tilt: non-price evidence (flows, COT week improvement, DRAM) strengthens the bullish case, but conviction remains moderate (0.36) because price has been partially realized and structural risks remain (COT still net-short, short-interest increases, upcoming macro prints). | +2.8% | 2026-07-29 |
| Metals Price & Demand metals_price_demand | SLV iShares Silver Trust | Long | 0.30 | Maintain a medium-term long bias on SLV (1–3 months) while staying tactical and risk-aware. Price-derived signals show a strong run: 42-day price_in excess +21.13%, arc-window excess +15.04%; the 42d etf window cumulative +21.13% with a max drawdown -10.19% (latest close 62.72), indicating sizable realized gains and volatility risk. Crucially, non-price, asset-specific evidence strengthens the precious-metals repricing thesis: CFTC COT (silver) net long +11,695 contracts (report 2026-08-18), at the 90th percentile, indicating continued speculative/hedging net long positioning; 10Y real yield = 2.35% (20d -8bp) — a recent decline in real rates lowers the opportunity cost of holding non‑yielding metals; prediction markets for gold are strongly bullish for August highs (polymarket yes_prob $4,600=1.00, $4,700=0.81), supporting cross-metal risk appetite. Flows are mixed: SLV 30d cumulative flows +1.02% AUM but net outflow -$27M (-0.08% AUM) since 2026-08-17. Options IV is elevated (ATM IV 45.58%, IV percentile 80%), signaling priced-in volatility. Net: non-price, direct signals (COT + real rates) reinforce the long thesis, so I upgrade conviction versus the prior weaken—but crowding (high COT pctile, high IV, price distance to MA50) constrains position sizing and overall confidence. | +16.7% | 2026-06-23 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | KBE SPDR S&P Bank ETF | Contested | 0.25 | Maintain the prior 'contested' view. Price has already priced some dovish Fed expectations in (price_in excess = -3.55%, KBE cum +7.05% vs XLF +10.60%; etf_window cum +7.05%, max drawdown -4.26%, latest close 68.74), but non-price, quantitative signals remain mixed and offsetting: Fed net liquidity (WALCL−TGA−RRP) is $5,792B with a 4-week change of -$125B (marginal tightening / mild hawkish pressure); prediction market probability that the Fed hikes in 2026 is 0.47 (no clear bias); KBE short interest 15,145,952 shares, days-to-cover 11.2, 7-period short accumulation +265% (crowding/squeeze constraint); flows since 2026-08-14 show net redemptions -$41M (-2.43% AUM) while 30d cumulative +4.60% AUM. The triggering event — a semiconductor startup poaching Nvidia talent — is company/sector-specific and lacks a plausible transmission channel to Fed policy or regional bank fundamentals. It therefore does not provide a valid non-price, arc-relevant signal to change direction. Keep contested and await core PCE / FOMC minutes or other clear non-price policy signals before taking a directional mid-term position. | +3.5% | 2026-06-07 |
| Russia-Ukraine War russia_ukraine_war | GLD SPDR Gold Shares | Long | 0.24 | Maintain a medium-term (1–3 month) long on GLD under the geopolitical-escalation→safe-haven thesis. Critique of prior: the 2026-08-18 note correctly downplayed conviction because price had partially priced in the move and rising 10y real rates were a headwind; that was prudent. New independent evidence since then, however, strengthens the structural case: CFTC COT net long rose from +137,662 to +141,648 contracts (report week 2026-08-18; WoW +3,986; OI=406,260 → continued and increasing long positioning), and fund-side real creations remain large (since 2026-08-03 net creations +$2,015M; 30d +2.01% AUM), indicating ongoing allocation into GLD independent of price action. Real 10y yields eased slightly (10y real rate 2.35%, 20d -4bp) reducing carrying-cost headwind. Price has indeed moved (arc price_in excess +1.22%; 8d etf window +3.04%, max DD -1.47%), and technicals/IV show crowding and hedging demand (RSI 71; dist to MA50 +10.55%; ATM IV 27.26%, 96th pctile). Because COT and ETF flow are independent non-price signals that map cleanly to the safe-haven transmission chain, I strengthen the long view modestly but cap conviction due to crowding/overbought signals. | -4.6% | 2026-06-04 |
| Private Credit & Shadow Banking Stress private_credit_shadow_banking_stress | KRE SPDR S&P Regional Banking ETF | Short | 0.22 | Recap of prior view (2026-08-18): I previously recommended short (confidence 0.12) based on two non-price, arc-relevant signals—KRE net flows / AUM contraction and a drop in Fed net liquidity—forming the chain “reduced funding supply → regional bank repricing/liquidity stress.” I flagged that short-term price/flow had briefly diverged (inflows since 8/10), so confidence was intentionally low. Current update: maintain short and strengthen. Independent, non-price quantitative evidence has re-confirmed the transmission channel: 1) KRE experienced net redemptions of -$81M (-1.86% AUM) since 2026-08-13 (data as of 2026-08-19); 30-day cumulative flow -1.15% AUM; 90-day -9.27% AUM — indicating renewed capital withdrawal from the ETF (a direct valuation/liquidity pressure channel). 2) System liquidity remains contracting: Fed net liquidity (WALCL−TGA−RRP) is $5,792B with a 4-week change of -$125B (as of 2026-08-19), supporting the macro channel “liquidity withdrawal → financial intermediation stress.” Price evidence is directionally consistent: arc price_in excess = -2.86% (KRE -0.35% vs XLF +2.51) and the 30-day etf window shows cumulative -0.35% with max drawdown -4.13%; treat these as already-reflected corroboration, not the primary driver. Contradictory signals: option IV sits at the 14th percentile and put-call skew is slightly negative vs XLF, implying options markets are not pricing acute stress. Historical-analog strength remains weak (many first_signals, unclear market_context), so position sizing should remain cautious. Net: keep short and raise confidence to 0.22 (within the +0.10 cap) because fresh non-price flows + renewed evidence of liquidity drawdown independently reinforce the original thesis. | -2.6% | 2026-07-12 |
| Korea / US Geopolitics korea_us_geopolitics | GLD SPDR Gold Shares | Long | 0.22 | Maintain a mid-term long on GLD with increased conviction. Non-price quantitative evidence has not only persisted but strengthened — CFTC COT net longs rose from +137,662 to +141,648 contracts (reporting week 2026-08-18, +3,986 week-on-week, at the 100th percentile), and ETF creations/redemptions show net inflows of +$4,529M since 2026-08-10 (~+2.94% AUM; 30-day +3.45% AUM). These are direct, asset-specific (gold) demand/signaling metrics supporting higher positioning. The macro backdrop is not hostile: 10y real yields sit at 2.35% (20d −8bp), a mild tailwind for non-yielding metals. Price has already moved (arc excess +3.80%; etf_window cumulative +3.8%, max drawdown −11.26%) and short-window historical analogs are mixed (some 8–9d +3%–+7% rallies vs prior 6–8d −2%–−4% pullbacks), indicating material short-term volatility and crowding risk. The current trigger (event #4166) is not Korea-specific and does not supply an arc-relevant causal link, so it does not overturn the non-price evidence. Net: stronger basis to keep a mid-term long position, but remain cautious because of record COT crowding, elevated IV (96th percentile), and technical/flow-driven crowding. | +2.3% | 2026-06-03 |
| AI Compute Capex Race ai_compute_capex_race | CHPS Defiance Semiconductor Equipment ETF | Long | 0.14 | The prior assessment (2026-08-19) correctly stayed contested: ETF price/flow weakness (arc cum CHPS ~-14.54%, excess vs SMH ~-3.86%, large drawdown) plus absent quantifiable capex/order data made a one-sided call premature. This refresh shifts to a conditional long because of independent, non-price evidence signaling supply tightness in memory — TrendForce DRAM spot prices are up materially (aggregate trend ~+9.16%; DDR5_16Gb +13.85%), which directly supports the causal chain (AI training/inference -> higher DRAM demand -> semiconductor/capex uplift). Price-derived metrics remain mixed (etf_window 9d cum -1.36%, max drawdown -7.42%; arc cum CHPS -15.42% vs SMH -12.15%, excess -3.27%), though short-term price_in excess is +0.92% (3d). Short interest fell ~18.1% (days-to-cover=1), reducing squeeze dynamics. In sum: DRAM spot strength is an independent non-price strengthening signal for the AI compute capex channel, justifying a lean-long posture for 1–3 months, but execution should remain event-driven because direct capex/order confirmations are still absent. | -16.9% | 2026-06-03 |
| US-China Tech Decoupling us_china_tech_decoupling | VDE Vanguard Energy ETF | Long | 0.12 | Prior assessment summary: On 2026-08-18 I stayed small‑long VDE with very low conviction (0.05, decaying), arguing price had already priced much in and there was little direct non‑price evidence tied to the arc's causal chain. Critique of prior: that note correctly flagged price exhaustion risk and the need for 'arc‑specific' evidence, but it underweighted continuing, independent supply/demand signals. Current view (still long): price has materially run (arc price_in_excess = +13.93%; arc cumulative VDE +16.80% vs SPY +2.87%), which reduces residual upside. However the fresh snapshot delivers independent, arc‑relevant non‑price signals: 1) WTI near/12m slope = +17.76% (backwardation), consistent with near‑term tightness and an upstream premium; 2) EU / German manufacturing PMI surprised to the upside (DE manufacturing PMI = 54.1 vs exp 52.1, σ=2.0), supporting demand resilience; 3) FINRA short interest remains elevated (shorts +40.5% vs prior, days_to_cover = 1), implying crowding/short‑squeeze risk. These pieces form a plausible causal chain (spot tightness & demand → producer premium → VDE outperformance) independent of price moves, so I modestly strengthen the small long stance (confidence from 0.05 → 0.12) but keep position size cautious because much of the move is already priced. Quantitative anchors cited: arc price_in_excess +13.93%; arc VDE +16.80% / SPY +2.87%; 6d etf_window cum +4.31% / max DD -0.08%; WTI slope +17.76%; DE PMI=54.1 (σ=2.0); options IV percentile 13%. | +13.9% | 2026-06-20 |
| Trump 2026 Tariff Policy trump_tariff_2026_policy | MCHI iShares MSCI China ETF | Contested | 0.12 | Maintain contested (do not take a directional long/short in this arc). The prior core judgment—that there is no direct USTR/White House tariff execution evidence and therefore one should not convert narrative heat into a trade—is still valid. Price has already partially priced this risk: price_in_excess (43d) = 11.04% (MCHI +5.3% vs EEM −5.74%), arc-window cumulative MCHI +4.16% / EEM −5.46% (excess +9.62%); the recent 9-day etf_price_window shows only +0.07% cumulative with max drawdown −2.16%, indicating no clean continuation signal. Crucially, the triggering item is a Fed/governance article (Kevin Warsh struggling), not a tariff implementation signal, so it does not provide the causal chain needed to trade the tariff arc. Non-price quantitative signals favor caution: Fed net liquidity WALCL−TGA−RRP = $5,792B (4-week Δ = −$125B — marginal liquidity withdrawal is a headwind for risk assets), VXEEM = 27.05 (z = −2.06 — volatility at a low relative level), and MCHI short interest = 15,353,069 (+19.1% over recent periods; days-to-cover = 5.94 — crowding/short-squeeze risk). Arc media heat has accelerated (5-day mentions 47 vs prior 17, 2.76x), increasing crowding risk. Decision: remain contested. Only a direct, non-price tariff execution signal (USTR/White House tariff list or matching trade/flow/sanction data) or a clear non-price driver for EM downside would move the arc to a directional trade. | +2.4% | 2026-06-20 |
| EU-US Tariff Dispute eu_us_tariff_dispute | LIT Global X Lithium & Battery Tech ETF | Long | 0.10 | Prior assessment was contested with very low confidence because it correctly highlighted the absence of independent non-price quantitative evidence and warned against relying solely on price moves. Since then LIT has continued to rally (arc window LIT cumulative +12.98%, excess vs MXI +2.47%, 5d excess +0.23%)—a partial price-in of information—but price alone is not a directional proof. The fresh, independent non-price signal is stronger EU manufacturing: eu_pmi_manufacturing=52.8 (σ=+1.2) and notably strong Germany manufacturing de_pmi_manufacturing=54.1 (σ=+2.0). Causal chain: stronger EU manufacturing → higher industrial demand for critical minerals / battery metals → supportive demand impulse for the lithium/critical-minerals complex and therefore LIT. Balancing risks: 30‑day AUM flow is negative (-1.77% AUM), short interest has fallen (696,192 shares, -27.1%, days-to-cover=3.07) reducing squeeze dynamics, and much of the move is already priced (etf_window cumulative +12.98%, max DD -2.91%). Conclusion: modest long tilt on LIT for the 1–3 month horizon, low conviction and position sizing, monitor upcoming macro prints as potential re‑rating triggers. | +9.3% | 2026-07-24 |
| Russia-Ukraine War russia_ukraine_war | IAU iShares Gold Trust | Long | 0.10 | Thesis: Within the Russia-Ukraine War arc, geopolitical risk remains a structural bullish driver for gold (escalation → safe‑haven demand → precious metals ETF flows). No new event joined this refresh, but independent non‑price quantitative evidence strengthens the long case vs the prior assessment: CFTC net long rose to +141,648 contracts (reporting week 2026-08-18, recent series climbing; at the 100th percentile), IAU experienced real creations of +$315M since 2026-08-17 (+0.46% AUM), and 10Y real yields eased to 2.35% (20d −8bp). These are direct non‑price signals supporting higher allocation to a non‑yielding asset. Offsetting considerations: IAU has already delivered excess return +6.65% over the arc window, the 8‑day ETF window is +3.04% (drawdown −1.45%), technicals are extended (RSI14=70.9) and IV sits at the 79th percentile — indicating crowding/追高 risk. Net: non‑price evidence independently strengthens the safe‑haven linkage so we upgrade to a low‑confidence long (confidence=0.10) while recognizing price has partially priced this in. | +3.1% | 2026-06-07 |
| East Asia Semiconductor Supply Chain east_asia_semi_supply_chain | AIQ Global X Artificial Intelligence & Technology ETF | Long | 0.08 | Tilt long on AIQ but keep conviction low. The prior assessment set contested (confidence 0.02) because AI compute capex and DRAM tightness were the primary non-price bull channels while flows/options/shorts produced mixed signals. No new event opps arrived since then, but the fresh snapshot provides a direct, non-price reinforcement: DRAM spot prices rose sharply (DDR5_16Gb $37.0 +13.85%, DDR4_16Gb $42.0 +4.48%, DDR4_8Gb $23.5 +16.92%), signaling memory supply tightness → supportive for AIQ’s memory/semiconductor holdings’ revenues and pricing power. Secondary support: AUM net inflow +$16M since 2026-08-11 and 30-day cumulative +1.08% AUM. Price context: arc-window cumulative AIQ -1.87% with AIQ vs QQQ excess +0.38% and max drawdown -16.56% — the fundamental tailwind appears only partially priced. Offsetting risks: option skew (put IV 32.18% vs call IV 28.81%, skew +3.37pt; +5.33pt vs QQQ), FINRA short interest +493.6% (days-to-cover 1.59), low VXN (z=-1.04), and Fed net liquidity -$125B (4-week) indicate positioning and liquidity/framing risks. Net: DRAM spot constitutes an independent, arc-relevant non-price reinforcement so we modestly strengthen the bull tilt, but keep conviction low and position sizing small given conflicting positioning/liquidity signals. | -3.9% | 2026-06-16 |
Nascent 70
| Narrative arc | ETF | Direction | Confidence | Thesis | Excess vs SPY | First seen |
|---|---|---|---|---|---|---|
| Private Credit & Shadow Banking Stress private_credit_shadow_banking_stress | PFF iShares Preferred and Income Securities ETF | Short | 0.72 | Maintain a short stance on PFF (1–3 month horizon). The prior assessment was short with 0.72 confidence (contested). Causal chain remains: private-credit / shadow-banking repricing → persistently elevated/widening high-yield spreads → fund outflows and growing short interest → mid-term price pressure. Fresh non-price data continue to support the repricing/fund withdrawal channel: HY OAS = 2.75% (5d +4bp), PFF AUM $13.1B with net outflows -$60M since 2026-08-14 (-0.46% AUM, 5 creations/redemptions), 30-day cumulative flows -1.72% AUM; FINRA short interest 5,049,497 (+11.3%), days-to-cover = 1.85. Price has not fully reflected the thesis: arc-window cumulative PFF -0.56% vs SPY -0.90% → PFF excess +0.34% (5d), etf_window cum -0.56%, max drawdown -0.59%, last close 30.46. Options IV is low (IV percentile 40%); skew is relatively less negative vs SPY (difference +0.33pt) but skew/IV are context, not primary directional drivers. Net: independent non-price signals still lean negative but are not materially stronger than in the prior snapshot, and price/IV do not confirm a panic repricing — therefore keep the prior short view at the same confidence (0.72) and status contested. Historical analog strength remains weak (first_signal); remain cautious ahead of upcoming US core PCE/GDP prints. | +0.3% | 2026-08-17 |
| RBA Australia Rate Cycle rba_australia_macro_policy | SPY SPDR S&P 500 ETF Trust | Long | 0.66 | Acknowledge and critique prior view: my prior moderate long (confidence=0.63, contested) relied on two non-price pillars—large net creations in SPY since 2026-08-13 (+$11,407M = +1.39% AUM) and an extremely crowded short position in S&P futures (CFTC net ≈ -280k contracts, ~98th percentile). I also noted the lack of a direct RBA causal chain then. Current judgment: remain long and modestly strengthen conviction. Structured quantitative support: (1) non-price funding/position signals remain constructive—SPY net inflows +$11,407M and CFTC net -281,402 contracts (98th pct) imply continued short-covering potential; (2) price cross-check shows a small pullback but no structural break: price_in excess (3d) = -0.43%, etf_price_window (9d) cumulative = -1.35% with max drawdown -1.96%; (3) a new, arc-relevant non-price datum strengthens the chain: Australia CPI is forecast to fall to 3.3% on 2026-08-26 (from prior 3.8%), which—if realized—would materially lower the expected RBA tightening path and support global risk assets via lower local and potentially cross-border rate expectations. Offsets: Fed-related net liquidity fell $125B over 4 weeks (WALCL−TGA−RRP = $5,792B, 4w -125B), a modest liquidity headwind; option IV remains low (IV percentile 24%), indicating cheap protection. Net: the non-price evidence (flows, CFTC) plus the Australia CPI forecast provide incremental structural support, so I modestly strengthen the long stance while keeping risk management tight. | +0.0% | 2026-08-19 |
| RBA Australia Rate Cycle rba_australia_macro_policy | VTI Vanguard Total Stock Market ETF | Long | 0.62 | Maintain a medium-term long on VTI. Rationale: the triggered blueprint remains macro_data_inflection (UK Q2 growth), which can mechanically translate into stronger global risk appetite → short-covering / incremental risk asset buying that favors broad US equity exposure. Non-price structural evidence: CFTC S&P net positions remain net-short but showed earlier partial covering and active positioning (four-period sequence: -297,476 → -329,999 → -280,446 → -281,402 contracts; latest net -281,402 with rising OI), the same class of signal I previously cited to argue for short-covering tailwinds. Price-derived cross-checks show the move has not been strongly reflected: price_in_excess = -0.03% (3d); etf_window cumulative = -0.46% with max drawdown -0.9% (3d). These are corroborative only per system rules. Options context: ATM IV is low (call 13.61% / put 11.09%, IV percentile 0%) and skew -2.52pt (relatively more expensive calls vs SPY), but skew alone is not a directional source. Key limitation: the event is not Australia-specific — the decisive RBA transmission point is AU CPI due 2026-08-26 — so the causal chain to RBA policy is indirect and historical-analog strength is weak (first_signal). Therefore keep long exposure, but with modest confidence (0.62) rather than a stronger tilt. | -0.0% | 2026-08-19 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | SOXX iShares Semiconductor ETF | Long | 0.60 | First evaluation: within the Fed 2026 rate-cycle arc, I lean long SOXX for the 1–3 month horizon based on a macro/earnings data inflection. Rationale: (1) The trigger is a macro-data/earnings inflection — widespread large-cap Q2 beats (FactSet: ~86% beating estimates; earnings ~29% above estimates, source: FactSet via article) which eases AI-spending and inflation concerns and reduces immediate risk-off repricing pressure. (2) Industry demand signal: DRAM spot prices are rising (TrendForce: aggregate +9.16%, DDR5 +13.85%), consistent with ongoing AI/data-center-driven memory demand. (3) Flows & risk appetite: SOXX had net creations of +$1,543M (+3.64% AUM) since 2026-08-14, indicating fresh allocation; VXN=23.26 (z=-1.04) signals low volatility/risk-on backdrop. (4) Rate-path signals do not point to imminent hawkish surprise: prediction market for a 2026 Fed hike shows yes_prob=0.47 (manifold), under 50%. Offsetting risks: SOXX 30/90d flows are -0.60% / -8.75% AUM (medium-term outflows), put-call skew is +2.01pt vs QQQ (greater demand for downside protection), and Fed net liquidity fell -$125B over 4 weeks (WALCL−TGA−RRP=$5,792B). Net: the earnings-driven macro-data inflection and memory-price demand signal support a tactical long, but position sizing should account for liquidity and hedging signals. | — | 2026-08-23 |
| NATO / EU Defense Spending Cycle nato_eu_defense_spending | VIXY ProShares VIX Short-Term Futures ETF | Long | 0.60 | First evaluation (no prior). Base view: take a tactical long on VIXY for the 1–3 month arc—conflict escalation tied to NATO/EU defense dynamics is a credible trigger for a safe-haven/volatility spike. Structural, non-price evidence: 1) CFTC COT shows a net short positioning of -19,093 contracts (0th percentile; wk change -6,966 or -57%), with OI=417,768 (+74,674 vs 2026-07-28) — a crowded short that is vulnerable to a rapid repricing; 2) VIXY has seen large net creations: +$54M inflow (+24.15% AUM) since 2026-08-10, signaling cash demand for volatility exposure which can amplify moves; 3) Option-side protection is cheap: ATM IV historical percentile = 0%, so IV can gap higher rapidly. Price cross-check: 5d excess return -2.34%; 9d window cumulative -5.65% (drawdown -5.65%) — recent price action is short-term bearish versus the thesis but per rules price alone does not overturn structural signals. Offsets: VIX term-structure is in contango (VIX/VIX3M = 0.866), and prediction markets assign low short-term NATO clash probabilities (e.g., Aug31 yes_prob=0.04). Net: structural signals (COT + AUM + cheap IV) imply asymmetric upside risk to volatility on an escalation path → tactical long VIXY with moderate-high conviction (0.60). | — | 2026-08-24 |
| Israel-Hamas Gaza War israel_hamas_gaza_war | GLD SPDR Gold Shares | Long | 0.60 | First assessment for the Israel-Hamas Gaza War arc tracking GLD: lean long (1–3 month horizon). Causal chain: the reported Iran-linked targeting of Kuwait increases the risk of regional spillover, lifting safe-haven demand. This is backed by non-price signals: CFTC net speculative long = 141,648 contracts and has risen in four consecutive reports (week-over-week +3%), indicating speculative positioning that amplifies bullish flows; real 10Y real rate is 2.35% with a 20-day decline (~-8bp), lowering the carry cost of gold; GLD has seen material creation flows (30-day net inflows ≈ +3.45% of AUM). Price behavior is consistent with the thesis (price_in 5d excess +5.31%; 9-day window cum +5.59% with max drawdown -1.71%), showing the market has started to price an escalation premium. Given the presence of direct non-price drivers (COT, real rates, fund flows) I favor a medium-strength long. Risk notes: overcrowding signs — GLD IV percentile 96%, RSI 71 and distance to MA50 +10.6%, and elevated single-day volume — mean a high risk of short-term mean reversion if escalation abates. Recommend phased entries and disciplined risk limits. | +1.8% | 2026-08-24 |
| EU-US Tariff Dispute eu_us_tariff_dispute | MXI iShares Global Materials ETF | Long | 0.58 | Maintain a modest medium-term long on MXI with slightly higher conviction versus the prior refresh. The prior thesis — Europe is not panicking about tariff escalation so materially lower odds of large tariff shock — still holds and is reinforced by fresh non-price evidence. Quantitative anchors: price_in excess = +7.47% (MXI +10.15% vs SPY +2.68%); etf_price_window cumulative +10.15% with max drawdown -4.93%, indicating price has partly reflected the thesis but without a collapse. Independent non-price signals: net ETF creations since 2026-08-17 = +$6M (+1.47% AUM) and intraday volume proxy $4.0M vs 20-day baseline $1.6M (2.36x, labeled bullish_surge) — consistent with active fund flows into MXI; EU/Germany manufacturing PMI surprised materially to the upside (DE manuf 54.1 vs exp 52.1, σ=2.0; EU manuf 52.8 vs exp 51.8, σ=1.2), supporting the causal chain (EU resilience → lower tariff escalation probability → relative support for MXI). Constraints: VIX term structure remains in contango (0.859), systemic liquidity (WALCL−TGA−RRP) 4-week change -$125B, and FINRA short interest +80.7% with days-to-cover = 2.98 — these cap position sizing and risk tolerance. Result: keep long, modestly stronger conviction (0.58), with disciplined sizing and watchlists for flow or liquidity reversals. | +7.5% | 2026-08-19 |
| RBA Australia Rate Cycle rba_australia_macro_policy | XLU Utilities Select Sector SPDR Fund | Short | 0.58 | I maintain and strengthen the mid-term short on XLU. Prior view (2026-08-19, short, confidence 0.48) rested on: 1) arc evidence of XLU underperformance (arc cumulative XLU -5.07% vs SPY +6.26%, excess -11.33%); 2) non-price signals implying macro resilience and liquidity contraction (previously cited EU ZEW/UK labor and Fed net liquidity drawdown); 3) historical analog (event_opp #40) where hawkish surprises led to XLU weakness (21d -4.62%). The current trigger (UK Q2 growth/FTSE) does not overturn this chain; it reinforces it with fresh non-price data: EU manufacturing PMI 52.8 (exp 51.8, σ=1.2) and DE manufacturing PMI 54.1 (exp 52.1, σ=2.0), indicating global macro resilience that lowers odds of near-term rate cuts. System liquidity remains contracting (WALCL−TGA−RRP $5,792B, 4w -$125B) and XLU has seen real outflows (since 2026-08-13 net -$165M, -0.73% AUM). Price signals: 75-day price_in shows XLU excess -13.56% (XLU -7.76% vs SPY +5.8%); the 9-day window is +1.48% (max dd -1.22%) — a short-term bounce but not a regime change. Historical analog (event_opp #40: 21d -4.62%, 6d excess -2.82%) is consistent. Given these combined signals I increase conviction from 0.48 to 0.58. Key caveats: the 9-day rebound and low option IV (26th pctile) can produce short squeezes; AU CPI on 2026-08-26 is a direct RBA-facing datapoint that could force reassessment if markedly dovish. | -13.6% | 2026-08-19 |
| NATO / EU Defense Spending Cycle nato_eu_defense_spending | BNDX Vanguard Total International Bond ETF | Long | 0.58 | Maintain a tactical long in BNDX (1–3 months). Chain: geopolitical/defense escalation → safe-haven demand → higher allocation into hedged international investment-grade sovereign/quasi-sovereign bonds (BNDX). Independent non-price evidence supporting this: strong Treasury auction demand (recent bid-to-cover mean ≈ 2.67, indicating institutional absorption of government debt), implied Fed-funds path priced near 3.725% (leaves room for market-rate easing expectations supportive for duration assets), and persistent tail-risk in prediction markets (polymarket yes_prob for NATO vs Russia by year-end ≈ 0.24). Price has only partially priced this (arc price_in excess = +0.13%; 3‑day cum +0.13%; max drawdown -0.19%; latest close 47.65), so gains are not fully realized. Technicals (MA20<MA50 death cross; RSI14=41.8) and sparse IV data are cautionary but do not overturn the structural thesis. Historical analog strength is low (first_signal), so remain position-sized prudently and monitor upcoming macro prints. Note elevated reported shorts (+53.8% vs prior, days-to-cover=1) as a crowding/volatility risk. Net: independent non-price signals reinforce the safe-haven demand channel, so I raise conviction modestly from 0.50 to 0.58 while keeping a prudent stance. | +0.3% | 2026-08-18 |
| East Asia Semiconductor Supply Chain east_asia_semi_supply_chain | EWT iShares MSCI Taiwan ETF | Long | 0.55 | Event summary: SK Hynix restarted construction on the Dalian NAND plant #2, targeting ~50% capacity uplift for the Dalian complex and production by H1 2027. The restart is ambiguous for the arc: it signals renewed AI/storage capex and regional industrial clustering (positive for the semiconductor ecosystem and upstream suppliers), but it is also a capacity expansion that will add NAND supply and may depress memory prices/profits once online. “Restart” is not an unambiguous demand signal. Quantified, structured evidence: - price_in / window: price shows modest positive reflection: 4d price_in excess = +0.14% (EWT -0.09% vs EEM -0.23%); arc window cumulative (2026-06-16→2026-08-17) EWT +3.86% / EEM -1.92% → excess +5.79%; window max drawdown = -19.83%. (Price is evidence of market reaction, not the sole direction driver.) - Non-price quantitative signals (directional drivers): • ETF flows: EWT 30-day cumulative net inflow +5.33% AUM; since 2026-08-14 net creations +$104M (+0.92% AUM) — capital allocation is supportive. • Options/positioning: call-side skew and OI tilt (put_call_oi_ratio ≈ 0.33; 30d skew_put_minus_call = -5.3; call_oi = 51,655) indicate market positioning is biased to the long side. • Short interest/positioning: short_shares = 8,062,620; days_to_cover = 1.19 — sizable short base with low cover days implies potential squeeze risk (crowding context). Causal chain used: restart → supply increase in NAND (near-term downward pressure on memory prices) but restart also evidences corporate/regional capex momentum that benefits the broader semiconductor ecosystem (foundry/fabless/equipment) which dominate EWT holdings. Given that non-price signals (flows and options positioning) are supportive and EWT’s composition benefits from AI capex, the balance for a 1–3 month horizon is mildly bullish (long), with only moderate confidence because of the supply risk and historical event mix skewed toward negative outcomes. Risk note: capacity expansions are not automatic buy signals — financing/expansion can dilute margins and increase supply; monitor memory price indicators and macro/FX that could flip the case. | -1.6% | 2026-08-23 |
| EU-US Tariff Dispute eu_us_tariff_dispute | SPY SPDR S&P 500 ETF Trust | Long | 0.55 | Acknowledging the prior view: on 2026-08-19 I recommended long (confidence 0.65) because the market reaction and non-price signals suggested the EU-US tariff shock probability had fallen and flows/volatility supported risk-on. Critical re-check: today’s trigger (event_opp #3502) is a sanctions/election item about Brazil, not a new EU→US tariff datapoint, so it does not reinforce the prior structural thesis. Price cross-check: the arc window has already priced a move—SPY arc cumulative (2026-07-01→2026-08-13) +4.31% (excess vs absolute +4.31%); the 36-day price_in/etf window shows cumulative +2.26% with max drawdown -3.38% and latest close 762.6. Non-price quantitative evidence still favors a modest long: SPY creation/redemption net inflows +$11,407M since 2026-08-13 (+1.39% AUM; 30-day cumulative +1.18% AUM), VIX 14.89 (z=-1.20), and CFTC COT net short -280,446 contracts but weekly change +49,553 (+15%) (recent short-covering). Countervailing signals: 5-day ATM IV +1.57pt, put-call skew movement, and Fed net liquidity down $125B over 4 weeks (WALCL−TGA−RRP = $5,792B) introduce increased uncertainty. Decision: keep long but reduce conviction (confidence → 0.55). Rationale: price and flows still support a medium-term long, but today’s event does not supply structural reinforcement for the tariff arc and liquidity/IV signals modestly weaken the case; with historical analogs remaining mostly 'first_signal' (weak analog strength) we calibrate lower conviction and recommend position sizing and event-driven stops. | +0.0% | 2026-08-19 |
| EU-US Tariff Dispute eu_us_tariff_dispute | FXI iShares China Large-Cap ETF | Long | 0.55 | Remain long FXI. Rationale: Prior view (2026-08-19, long 0.60) argued that a "shrug off" by EU leaders reduces the odds of an EU–US tariff escalation, easing trade-risk premia and favoring China/emerging-market equities. I explicitly critique the previous assessment: it leaned heavily on price-based evidence (price_in excess and window returns) and, while it did cite structural signals (short accumulation +72.4% and modest outflows), price should not be the sole directional driver. Reassessing current data, I keep the long stance but trim confidence. Supporting points with quantification: 1) price_in confirmation — 36d price_in excess = 11.36% (FXI 11.57% vs EEM 0.21%); arc-window cumulative (2026-07-01→2026-08-13) FXI +9.04% / EEM +0.30% => excess +8.74% (used only as tape cross-check); 2) non-price structural signals remain supportive: short interest cumulative +72.4% (days-to-cover = 3.25) indicating crowded shorts/short-squeeze potential; 30d net flows = -1.09% AUM (small outflow, not a large exit); VXEEM = 26.54 (z = -2.27) shows low emerging-market volatility / market complacency; options: ATM IV low (call IV 20.85%, put IV 17.77%, IV percentile 0%) and put-call skew -3.08pt (cheaper protection / relatively higher call demand). Historical analogs for tariff_policy_shock events often show short-window positive FXI returns (e.g., #2372 +6.01% in 8d, #2779 +5.80% in 8d, #2613 +2.52% in 10d). Counterpoints: the price run already carries much of the upside and short crowding raises downside if an adverse macro/shock occurs. Crucially, today's triggering event (#3502) concerns Brazil/sanctions and is not a direct, arc-specific non-price driver of an EU–US tariff rollback. Therefore maintain long but marginally lower confidence (0.55). | +9.5% | 2026-08-19 |
| RBA Australia Rate Cycle rba_australia_macro_policy | VOO Vanguard S&P 500 ETF | Long | 0.53 | Maintain a mid-term long on VOO. The prior assessment (2026-08-21) constrained the decision to non-price, RBA-relevant evidence and supported long using extreme CFTC net-short positioning, elevated systemic liquidity, narrow credit spreads and a contango VIX term structure. I critically rechecked those anchors: they remain intact and not contradicted. VOO shows a small recent price drag (price_in excess = -0.45% over the 3-day window; etf_price_window cumulative = -0.45%; max drawdown = -0.83%), but price is only a tape cross-check per rules and cannot by itself flip the thesis. The triggering news (FTSE/mining + UK Q2 growth) is not causally linked to the RBA rate path, so it is neutral for this arc. Therefore, structural non-price signals continue to favor risk assets → S&P500 → VOO; no Australia-specific macro or RBA communication has appeared to alter the RBA path. Keep long, contested, with a slight confidence reduction to account for modest recent excess weakness and very low option IV (market complacency). | -0.0% | 2026-08-19 |
| EU-US Tariff Dispute eu_us_tariff_dispute | XLB Materials Select Sector SPDR Fund | Long | 0.50 | Maintain a medium-term (1–3 month) long bias on XLB but with reduced confidence. The prior structural thesis — that calmer EU leadership responses reduce tariff_policy_shock probability → lower sector risk premium → supportive for materials — remains intact because today’s trigger is unrelated to the EU‑US tariff dispute (the event is Brazil election/sanctions coverage). Non‑price evidence still partially supports long: EU ZEW 31.4 vs exp 25.9 (2026-08-18) suggests EU sentiment is not being knocked by trade fears; option IV is low (ATM call IV 20.56%, put IV 17.92%, IV percentile ≈0%), indicating limited risk premia for the sector. Price checks do not indicate a breakdown: price_in (36d) excess = +0.49% (XLB 2.74% vs SPY 2.26%); etf_price_window (36d) cumulative +2.74% with max drawdown -3.81% and latest close 52.42. However, confidence is reduced because of countervailing non‑price flows and positioning: net creations/redemptions since 2026‑08‑13 = -$44M (-0.51% AUM) and 30d cumulative flows -0.38% AUM (funds marginally withdrawing), while FINRA short shares rose +8.8% vs prior (near‑term cumulative +19.3%, days‑to‑cover 1.37), implying crowding and potential squeeze/volatility risk. Therefore retain the long view but moderate the conviction (from 0.60 → 0.50). | +2.3% | 2026-08-19 |
| EU-US Tariff Dispute eu_us_tariff_dispute | EEM iShares MSCI Emerging Markets ETF | Long | 0.50 | Maintain the prior medium-term (1–3 month) long stance on EEM. The prior causal chain—EU leadership unlikely to escalate into large-scale retaliatory tariffs, reducing downside risk to EM exports/growth and benefiting trade-sensitive EM equities—has not been invalidated by new evidence. Non-price quantitative signals continue to support risk exposure: net creations/redemptions since 2026-08-14 net +$480M (~+1.56% AUM), and VXEEM=26.54 with z=-2.27 indicating a comparatively low-volatility, risk-friendly backdrop. Price and relative performance remain weak (arc window EEM +0.30% vs SPY +4.31%; EEM–SPY excess = -4.01%; 36-day price_in_excess = -2.05%; etf window max drawdown = -9.62%), but per rules these price-derived metrics are cross-checks rather than direction drivers. Countervailing signals — rising short interest (74,906,755 shares, +3.8%, days-to-cover=2.98) and a relatively less negative option skew (EEM skew -0.90pt vs SPY -2.10pt) — limit confidence. Conclusion: keep the long view but with slightly reduced conviction (confidence 0.50 versus prior 0.55) and await arc-specific structural confirmations (e.g., EU policy statements or stronger EM demand data) before scaling exposure. | -1.7% | 2026-08-19 |
| EU-US Tariff Dispute eu_us_tariff_dispute | VTI Vanguard Total Stock Market ETF | Long | 0.48 | Recap of prior view (2026-08-19): I previously recommended long (confidence 0.55). The rationale then was that EU leaders' calm stance reduced the instantaneous probability of a large tariff shock; supporting non-price evidence included CFTC SP500 positioning (net moved toward less negative) and an outperformance in EU sentiment (ZEW), implying lower probability of policy escalation and a narrower risk premium for broad US equities (VTI). Price-derived data (arc window VTI ≈ +4% vs SPY ≈ +4.3%, VTI-SPY excess ≈ −0.24%) were treated as background. Current reassessment: I keep a long bias but downgrade conviction to 0.48. The primary non-price supportive signal—CFTC positioning—remains: net = −280,446 with a weekly change of +49,553 (+15%), consistent with short covering and a risk-on tilt. Price inputs show VTI has not meaningfully outperformed (36d price_in_excess = −0.28%; arc_window cumulative VTI +4.07% vs SPY +4.31%, excess −0.24%; etf_window cumulative +1.98%, max drawdown −3.29%), indicating the market has only partially priced any policy easing. I lower confidence because: (1) the current trigger (event_opp #3502) is a Financial Times item about Brazil's election / sanctions and is not causally connected to the EU–US tariff dispute arc, so it does not reinforce the prior causal chain; (2) the historical analogs for this arc are mixed-to-lean-bearish (sampled event_opps: long 10 vs short 14), reducing analog strength; (3) aggregate liquidity has declined modestly (Fed net liquidity $5,792B, 4w −$125B), a modest headwind for risk assets. Net: keep the prior directional view but weaken conviction—COT short-covering is the core non-price support, prices only partially confirm, and new event is neutral-to-irrelevant. | -0.2% | 2026-08-19 |
| China Financial Regulation china_financial_regulation | EEM iShares MSCI Emerging Markets ETF | Long | 0.45 | Prior assessment was contested (confidence 0.35), arguing offsetting non‑price signals: PBoC’s earlier ‘zero 7‑day injections + short‑term tool substitution’ looked like short‑end tightening, while broker recapitalizations, EEM net creations (+3.20% AUM 30/90d) and low VXEEM (z=-2.06) supported continued buying. I critique the prior’s persistence claim about the zero‑injection regime: the new, observable non‑price event—PBoC restarted 7‑day reverse repos with a fixed‑rate, quantity tender for CNY 950 bn on 2026‑08‑21—changes the operational stance on short‑end liquidity from “sustained zero 7‑day supply” to “active, on‑demand 7‑day supply.” Anchoring direction on non‑price quant evidence per system rules, the repo restart (950 bn), ongoing ETF net inflows (EEM +3.20% AUM; recent +$480M ≈ +1.54% AUM), and low VXEEM (z=-2.06) form a coherent signal that short‑term liquidity is being backstopped and supports a mild-to-moderate bullish tilt for EM/China equity exposure over the next 1–3 months. Price‑derived signals are mixed: arc window EEM -1.34% vs SPY +2.96% (EEM‑SPY excess -4.30%) shows relative weakness, while the 33‑day etf_price_window cumulative +1.34% (max drawdown -8.71%) and recent 5d price_in_excess +0.60% partially reflect the ambiguity. Given the direct central‑bank liquidity action and corroborating net inflows, I flip to long but keep confidence moderate (0.45) because elevated short interest (74,906,755, +3.8%, days‑to‑cover 2.98) and the prior relative underperformance remain material downside risks. | -1.4% | 2026-08-21 |
| US EV Subsidy & Buildout Cycle us_ev_subsidy_buildout | KARS | Short | 0.40 | Maintain short. The prior case relied on a corporate signal (Samsung SDI buying GM’s 49.99% stake in SynergyCells and pivoting the New Carlisle plant to energy storage/high-tech use; event_opp #3843) interpreted as a near-term negative for US passenger EV demand and the tempo of localized large-scale battery expansion (causal chain: corporate repositioning → reduced upstream/vehicle orders and buildout pace → compresses KARS’ supply-chain capture). The prior analysis also cited US retail sales -0.6% mom (2026-08-14, σ=-2.8) as quantitative evidence of weaker consumer demand and KARS FINRA short interest 42,100 (+1109%) days-to-cover=3.5 as positioning context. In this daily refresh, no independent non-price data reverses that direction: KARS options IV is missing (IV chain_quality=sparse, unreliable), there are no new corporate/policy events, and recent PMI prints are European-centric and do not directly rebut US consumer weakness. Therefore the original short thesis stands. Caveats: (1) price_in / etf_window is missing (cannot confirm whether the market has already priced in the thesis), (2) short interest remains crowded (squeeze risk), (3) US core PCE and GDP releases on 2026-08-26 are near-term event risk—reduce sizing and enforce stops around those releases. | — | 2026-08-16 |
| US-China Tech Decoupling us_china_tech_decoupling | XLB Materials Select Sector SPDR Fund | Long | 0.40 | Maintain a medium-term (1–3 month) long bias to XLB but with reduced conviction. The original causal chain remains intact: the announced 15% tariffs on polysilicon/solar components (effective 2026-12-04) should shift demand/pricing power toward U.S. upstream materials producers and benefit XLB constituents (tariff → reshoring/substitution → improved pricing/margins for domestic materials). However, independent non-price flow data since the prior note weaken the funding-side support: the earlier-cited net creation (+$24M since 2026-08-12) is no longer present — recent creation/redemption data show net outflows of -$44M (-0.51% AUM) since 2026-08-13 and 30-day cumulative flows of -1.80% AUM, indicating investors have been withdrawing rather than adding exposure. Short interest has continued to accumulate (15,040,364 shares, +8.8% vs prior, 7-period cumulative +19.3%, days-to-cover 1.37), creating both downside pressure and squeeze risk. Price partially realized the thesis: after an early post-announcement dip (8d window cum -2.04%, max drawdown -2.74%), XLB has rebounded on 2026-08-20→21 (+2.14%, excess vs SPY +1.73%), showing some re-pricing. Options IV sits at low levels (call IV 20.17%, put IV 17.04%, IV percentile 0%), signaling cheap protection. Net: the tariff-driven structural bull case still exists, but recent fund outflows and rising short interest are independent non-price signals that weaken the prior confidence — lower position sizing and await clearer flow/earnings/implementation evidence. | +1.7% | 2026-08-20 |
| Korea / US Geopolitics korea_us_geopolitics | VGIT Vanguard Intermediate-Term Treasury ETF | Long | 0.40 | Maintain a medium-term long on VGIT (1–3 months). The prior thesis anchored direction on quantified non-price signals: the market-implied Fed-path shows a mild dovish tilt (implied fed funds ≈3.72–3.735%), recent Treasury auction demand is strong (7y bid-to-cover ~2.67 average), and Treasury volatility is relatively low (MOVE ≈73). Price structure corroborates intermediate-duration resilience: arc-window excess = +2.68% (VGIT cumulative -0.36% vs TLT -3.04%) with a 52-day max drawdown of -1.59%, indicating the intermediate bucket has outperformed the long end but without a large outright rally. Countervailing evidence remains material: CFTC 5y COT shows sizable net short positioning (net = -2,169,814 contracts, ~40th percentile), and long yields/term premium remain elevated (10Y ≈4.69%, term premium ≈0.84%), posing upside rate risk. The triggering item today (Carson Block / passive investing critique) is a broad market-structure narrative and does not provide Korea-specific, non-price quantitative evidence to strengthen the arc ‘Korean peninsula escalation → safe-haven → VGIT’. Given the continued presence of non-price support for the long case and persistent but non-decisive headwinds, I keep the long stance at confidence 0.40, while monitoring CFTC flows and upcoming US macro (core PCE, GDP) that can reprice the Fed path. | -3.9% | 2026-08-17 |
| US-China Tech Decoupling us_china_tech_decoupling | EWG iShares MSCI Germany ETF | Long | 0.40 | Maintain a medium-term long on EWG, but with reduced conviction. Since the prior update (2026-08-18) there is no new arc-specific event that would reverse the thesis; price-derived metrics remain positive: price_in excess = +3.08% (EWG window cumulative +5.73% vs EFA +2.65%), window cumulative +5.73% and max drawdown −4.85%, indicating realized excess return with controlled drawdown. Non-price signals are mixed: flows remain supportive (30‑day net flow +5.29% AUM, AUM $1.7B stable since 2026-08-14) and lagged short-interest data show a −23.9% change (settlement 2026-07-31, days-to-cover 2.95), suggesting some short-covering/positioning buffer. Offsetting this, options-implied data show independent downside demand: ATM IV (call 17.77% / put 16.99%), IV percentile 67%, 5‑day IV +1.61pt, and a skew differential vs EFA of +2.53pt — an independent (non-price) sign of increased protection buying and softer risk appetite for this ETF. Historical analogs for the sovereign‑sanctions blueprint are mostly labeled first_signal/unclear, so analog strength is limited. In sum, no new arc-specific evidence to flip direction, but the rise in IV/protection demand plus partial price realization justifies lowering conviction from 0.50 to 0.40. | +3.5% | 2026-08-15 |
| US-China Tech Decoupling us_china_tech_decoupling | XLE Energy Select Sector SPDR Fund | Long | 0.40 | Maintain a medium‑term (1–3 month) constructive view on XLE. The structural transmission chain remains intact: WTI front‑vs‑12mo curve is still in backwardation (slope ≈ +17.962%), and CFTC speculative positioning remains net long at +79,916 contracts (recent sequence +63,979 → +92,943 → +86,958 → +79,916). These non‑price signals point to near‑term physical tightness and positive roll that support energy equity earnings/valuations and thus XLE. Critique of the prior view (2026‑08‑18): that assessment was directionally correct but too high in conviction — since then the thesis has been partially price‑realized. Arc cumulative returns show XLE +14.45% vs SPY +3.39% (XLE‑SPY excess +11.07%), and the 6‑day etf_price_window is +4.18% (max drawdown -0.16%), indicating much of the expected move is in the tape. Non‑price signals now give mixed support: ETF flows have seen net outflows since 2026‑08‑13 of -$168M (-0.40% AUM) and 30‑day cumulative -1.41% AUM, and CFTC week‑over‑week shows a modest unwind (-7,042 contracts, -8%). Therefore I reduce conviction (0.48 → 0.40) rather than flip direction: structure supports long, but realization + outflows and slight positioning unwind weaken the trade. | +10.4% | 2026-08-15 |
| RBA Australia Rate Cycle rba_australia_macro_policy | IEF iShares 7-10 Year Treasury Bond ETF | Contested | 0.38 | Remain contested: no RBA/Australia CPI or policy trigger has arrived, so there is no clean causal path to force a unilateral duration view in IEF. Price and non-price metrics are mixed: price_in_excess (IEF vs AGG, 3d) = -0.12% (IEF mildly underperforming), etf_window (9d) cumulative +0.26%, max drawdown -0.49% — no clear price realization. Non-price evidence conflicts: CFTC 10y net = -2,229,013 contracts (net short, at the 3rd percentile) and IEF 30d net flows = -9.67% AUM point toward duration weakness, while implied Fed funds = 3.74% (slightly higher than previously) reduces aggressive cut bets. Offsetting support comes from low MOVE = 73.4 and strong auction bid-to-cover (~2.67). Overall, positioning/flows tilt mildly negative but market pricing and auction demand support long-end; signals cancel out, so keep contested and await arc-specific Australian data/policy to resolve direction. | -0.2% | 2026-08-19 |
| US-China Tech Decoupling us_china_tech_decoupling | KWEB KraneShares CSI China Internet ETF | Short | 0.38 | Maintain a short bias on KWEB (moderately low confidence). The prior view rested on two non-price chains: arc-specific regulatory/repricing risk (regulatory_action_chinese_tech) and a monetary/liquidity amplification path (macro_data_inflection), which were reinforced by the FOMC minutes and a 4‑week systemic liquidity decline of −$125B. On review the non-price quantitative evidence still supports a downside transmission: the FOMC minutes continue to signal that inflation remains elevated (raising policy-tightening risk), WALCL−TGA−RRP 4‑week change = −$125B (reducing liquidity), and FINRA short interest = 41,541,170 (7‑period +16.4%, days‑to‑cover = 1.93) indicates market positioning is biased toward downside but crowded. Price-structured measures partly validate short exposure: etf_price_window (9d) cumulative = −5.09% (max drawdown −5.09%); price_in (42d) excess = −0.84% (KWEB 6.43% vs FXI 7.27%) — i.e., recent price action has not disproved the thesis. Offsetting constraints: option IV at the 9th percentile (cheap protection), VIX term-structure ratio = 0.866 (calm contango), and arc-window cumulative KWEB +6.91% vs FXI +4.88% (excess +2.03%) imply residual relative strength and squeeze risk. Net: the non-price causal chain remains intact and recent short-term price moves are consistent with the short thesis, so we keep short with moderately low confidence due to crowding and low IV. | +3.6% | 2026-08-14 |
| Latin America Economy latin_america_economy | XLF Financial Select Sector SPDR Fund | Short | 0.36 | Maintain short. The prior causal chain — weaker macro (retail sales surprise) → credit/liquidity stress → fund outflows from financials → earnings/valuation pressure on banks/financials — remains intact. The fresh non-price metrics continue to map onto that chain: XLF AUM $55.5B; net redemptions since 2026-08-14 of -$1,795M (~-3.23% AUM) and 30-day cumulative flow -2.22% AUM, indicating ongoing capital departure from the sector; HY OAS ≈2.75% and Fed net liquidity (4-week) -$125B indicate persistent credit/liquidity tightness. These non-price signals support the original downside transmission. Price evidence is mixed/contextual: arc window (2026-08-10→08-21) XLF cumulative -0.57% vs SPY -0.95% (XLF-SPY excess +0.37%); the 9-day etf window cumulative is -0.55% with max drawdown -2.25%. Price has not decisively realized the thesis nor contradicted the structural evidence, so confidence is not increased. Options IV sits at a lowish historical percentile (36%) with 5-day IV +1.8pt — useful context but not directional on its own. Net: non-price evidence persists and aligns with the short thesis, so we keep the prior stance with modestly reduced conviction (0.36). | +0.4% | 2026-08-09 |
| RBA Australia Rate Cycle rba_australia_macro_policy | VGIT Vanguard Intermediate-Term Treasury ETF | Contested | 0.36 | Maintain 'contested'. The core conflict from the prior note (2026-08-22 10:40) persists: CFTC 5y positions remain heavily net-short (net = -2,169,814 contracts; 4-period series -2,108,638→-2,211,439→-2,147,744→-2,169,814; 40th percentile), which is consistent with exposure favoring higher rates / duration pain. Offsetting that, market pricing and funding signals continue to support duration: implied Fed‑funds path ≈ 3.735% (still implying market-priced easing bias relative to current), MOVE = 73.4 (low realized/anticipated bond volatility), and the curve shows 2s10s = 0.5 with 10Y = 4.69% and term premium ≈ 0.8393% (signal that steepening is driven by long-end/term-premium rather than front-end hikes). Price tape shows VGIT price_in excess = +0.79% over TLT (VGIT 3d cum -0.38%, max DD -0.38%), i.e., relative resilience, but per rules price is only cross-check. Technicals (RSI14=44.8, MA20<MA50 death cross) are mildly weak. Options IV is missing, so cannot provide independent non-price signal; volume is above average ($289.0M vs 20d $188.4M, 1.54x) but only contextual. Near-term US core PCE/GDP events are scheduled in ~3 days. No new arc-specific non-price evidence resolves the contradiction; therefore remain contested with a slight downshift in conviction for time decay and technical/IV uncertainties. | +0.1% | 2026-08-19 |
| RBA Australia Rate Cycle rba_australia_macro_policy | KRE SPDR S&P Regional Banking ETF | Contested | 0.35 | Maintain prior 'contested'. The prior position correctly required that any directional call be grounded in Australia/RBA non-price quantitative evidence (e.g., a materially surprising Australian CPI, explicit RBA message shifting the path of rates, AUD/interest-spread moves tied to RBA guidance). Since the prior note there has been no Australia-specific non-price signal. The triggering item this cycle relates to UK mining/UK GDP and has no direct causal chain to RBA -> AUD/financial conditions -> KRE, so it does not change the arc. Price/flow facts: price_in excess (3d KRE vs XLF) = -0.19%; etf_window 3d cumulative = -0.19%, max drawdown = -0.39%, last close 74.86 (2026-08-21). Non-price context: KRE AUM $4.1B (2026-08-20) with net outflows since 2026-08-14 of -$427M (-10.51% AUM); 30d -8.69% AUM; 90d -17.45% AUM — these show investor withdrawal but are ETF-specific demand signals, not Australia-policy evidence. Option IV is low (call IV 23.82% / put IV 20.56%; IV percentile 14%), indicating cheap protection. FINRA short interest 50,746,684; days-to-cover 3.73. None of the fresh non-price signals provide the causal link from RBA/Australia macro to KRE valuation needed to resolve the arc; remain contested and await Australia CPI (2026-08-26) or direct RBA signaling. | +0.2% | 2026-08-19 |
| Korea / US Geopolitics korea_us_geopolitics | VIXY ProShares VIX Short-Term Futures ETF | Contested | 0.35 | First evaluation (no prior). For the Korea–US geopolitics arc and VIXY, the stance is contested — evidence is mixed. Non-price signals supporting a long exposure: large recent net creations into VIXY (+$54M since 2026-08-10, +24.15% AUM) and 30-day cumulative inflows (+19.52% AUM) indicate demand for volatility hedges. Option market activity shows elevated ATM IV (call IV 50.39%, put IV 51.37%) and notable call activity intraday, consistent with protective positioning. Non-price signals opposing long: CFTC COT for VIX net = -19,093 (week -6,966), at the 0th percentile → speculative positioning is net short; VIX term-structure ratio VIX/VIX3M = 0.866 (contango) and VIXY IV historical percentile = 0% — the market is not pricing acute tail-risk. Price/technical corroboration: VIXY 9-day cumulative return -5.65% (9d avg -0.78%), 5d excess return -2.34% (VIXY -3.24% vs benchmark -0.90%), technicals show downward momentum. Prediction markets assign a very low probability to a DPRK invasion (polymarket yes_prob=0.03). The triggering event text is weak and does not contain a clear, credible escalation trigger. Because flows/positioning and option/term-structure data point in opposite directions, and absent a concrete escalation event that would change COT/term-structure/IV materially, I treat the signal as contested and recommend avoiding a large directional VIXY position while monitoring for a definitive non-price shift (e.g., sudden COT flip to net long, term structure moving toward backwardation, or a sharp rise in prediction-market probabilities). | — | 2026-08-24 |
| BoE UK Gilt Cycle boe_uk_gilt_cycle | XLU Utilities Select Sector SPDR Fund | Contested | 0.35 | Keep a 'contested' stance. Prior thesis (2026-08-22) framed the causal chain: UK/BoE fundamentals → gilt yields and global core rates → higher discount rates → pressure on duration-sensitive utilities (XLU). New quantitative facts: price-derived signals show a 5-day XLU excess return of -2.29%, a 9-day cumulative return of -1.97% and max drawdown -3.48%; flows since 2026-08-14 are net redemptions of -$189M (-0.84% AUM), 30-day cumulative -0.17% AUM; options IV is at the 26th percentile and put-call skew = -5.44pt (more negative vs SPY). Critically, we cite non-price quantitative evidence: GB services PMI 52.8 (actual vs exp 51.8, σ=+0.5) indicating services resilience and potential BoE hawkish bias, and Fed net liquidity 4-week change -$125B (as of 2026-08-19) indicating marginal global liquidity withdrawal — both support an environment where rates could move up and hurt high-duration sectors. These non-price signals, together with modest outflows and short-window negative excess returns, reinforce the duration-pain channel; but contradictory inputs remain (the triggering media story forecasting UK slowing, an internal analog on 8/21 showing +2.06% over 8d, low IV percentile implying inexpensive protection). Net: non-price evidence modestly tilts toward downside for XLU, but mixed/insufficient to move from contested to a committed short, so we keep contested with slightly higher lean toward rate-driven pressure. | -2.3% | 2026-08-15 |
| China Rare Earth Export Restriction china_rare_earth_export_restriction | EWT iShares MSCI Taiwan ETF | Contested | 0.35 | Keep contested. Reviewing the prior judgment (2026-08-23): I previously labeled the arc contested (confidence 0.35) because China's small-batch approvals for Nvidia H200 chips constituted a direct counterexample to the 'export-control-driven Taiwan outperformance (EWT long)' thesis, and non-price signals were mixed (persistent fund inflows vs. rising short interest; option IV low / skew not a standalone directional signal). Self-critique and re-check for this refresh: 1) The historical reference used previously (China approving small H200 shipments) still holds — this trigger (event_opp #4161) reaffirms the policy relaxation and there is no evidence of a large-scale policy reversal; 2) Price action does not vindicate a clear one-sided view: arc-window EWT +1.77% vs EEM +2.36%, excess = -0.60%; 28-day max drawdown -12.76%, implying the market has not priced a strong export-control premium for Taiwan; 3) Non-price quantitative signals remain mixed and continue to support a contested call: net creations/flows since 2026-08-17 +$104M (+0.92% AUM; 30/90d +5.35% AUM) indicate demand, while FINRA short interest at 8,062,620 shares (+37.6%) with days-to-cover=1.19 shows accumulated shorts and attendant uncertainty; option IV is at a low historical percentile (8%), and skew is not a permitted standalone directional input. Conclusion: the new event reconfirms the policy-relaxation counterargument rather than overturning it, and no arc-specific, scaleable non-price evidence has appeared to convert contested → directional, so I keep contested at prior confidence. | +0.3% | 2026-08-21 |
| RBA Australia Rate Cycle rba_australia_macro_policy | IWM iShares Russell 2000 ETF | Contested | 0.34 | Keep contested. The prior assessment correctly argued that absent Australia-specific, non-price causal evidence (e.g., Australia CPI surprises, RBA statements or Australia-US rate spread moves), we cannot ascribe IWM moves to an RBA rate-path inflection. Fresh data: price_in excess (3d IWM vs SPY) = -0.15%; etf_price_window (9d) cumulative = -0.77%, max drawdown = -2.43% — indicating modest relative weakness in small caps. Non-price signals: CFTC COT (Russell_2000) net = -99,786 contracts (four-period trend more short, week Δ = -4,628, percentile = 0), and IWM AP/creation flows since 2026-08-17 = -$375M (-0.46% AUM) while 30d = +0.65% AUM. Options IV low (IV percentile 24%) but skew vs SPY shows greater downside premia (~+1.08pt). These structure signals bias cautious/negative on small caps but do not establish a causal RBA→IWM link. Therefore maintain contested. Price has partially realized the earlier “relative softness” observation, so confidence is trimmed modestly; no new Australia-specific non-price evidence has appeared to convert the arc to a directional call. Australia CPI on 2026-08-26 is a near-term potential trigger but not yet observed. | -0.1% | 2026-08-19 |
| NATO / EU Defense Spending Cycle nato_eu_defense_spending | VGIT Vanguard Intermediate-Term Treasury ETF | Contested | 0.33 | Maintain 'contested'. The prior assessment identified two opposing transmission channels: geopolitical escalation → safe-haven bid (supporting 5y-centered ETFs like VGIT) versus NATO/EU defense spending → increased medium-term sovereign issuance (pressuring VGIT). This daily-refresh shows no decisive non-price evidence to break the tie. Price action: arc price_in excess = -0.61% (VGIT cumulative -0.14%, max drawdown -0.38%) indicating modest relative weakness consistent with supply/pricing pressure, but magnitude is small and not sufficient alone to flip structural view. Non-price signals remain mixed and offsetting: CFTC 5y net position remains net-short at -2,169,814 contracts (weekly Δ -22,070, at the 40th percentile), suggesting supply/short pressure; Treasury auction bid-to-cover ~2.67 (2-auction mean) signals robust demand supporting duration; implied Fed-funds path 3.735% and MOVE = 73.4 (-2.95%) point to a calm, slightly dovish-priced bond market; prediction markets show very low probability of NATO-Russia clash (Polymarket yes_prob = 0.04), weakening the pure safe-haven channel. Historical analog strength remains 'first_signal' — insufficient to commit long/short. If future non-price evidence decisively tilts (e.g., COT reversal >100k contracts or auction demand collapsing b/c <2.0), reassess direction. | +0.1% | 2026-08-18 |
| Latin America Economy latin_america_economy | INDA iShares MSCI India ETF | Contested | 0.32 | Keep contested: the prior contested call (confidence 0.36) was appropriate — there remains no consistent, independent non-price quantitative evidence to move to a clear long or short. Price-in metrics show INDA underperformance: price_in excess = -2.32% (INDA +1.74% vs EEM +4.06%), etf_window cumulative +1.74% with max drawdown -2.62% — prices have not delivered a one-sided outcome. Non-price signals remain mixed: 30d AUM flows = -2.56% (90d -3.65%) indicating mid-term outflows; short interest +18.6% with days-to-cover = 4.09 (more shorts but higher squeeze risk); option put-call skew = +1.44pt vs EEM -2.66pt (delta +4.10pt) implying relatively greater demand for downside protection, while absolute IV is low (8th percentile) and 5d IV change is modest (+0.33pt). VXEEM = 27.05 (z = -2.06) signals a low-vol regime. The recent 'Election trail' / Brazil campaign reportage increases political narrative risk but lacks a measurable causal chain (e.g., FX intervention, capital controls, sovereign stress, or large quant flows) to flip the arc. Hence signals remain conflicted — some non-price items tilt mildly bearish, others (low IV, low vol, squeeze risk) constrain conviction — so maintain contested. | -0.5% | 2026-08-14 |
| RBA Australia Rate Cycle rba_australia_macro_policy | XLRE Real Estate Select Sector SPDR Fund | Short | 0.30 | Maintain a mid-term short tilt on XLRE but with reduced conviction. The prior assessment (short, confidence 0.35) rested on three non-price quantitative pillars: 1) real fund flows (XLRE net redemptions of -$155M since 2026-08-12, ~-1.82% AUM); 2) systemic liquidity contraction (Fed net liquidity 4-week change -$125B as of 2026-08-19); 3) accumulation of short interest (FINRA short shares 7,454,300, +4.6%, days-to-cover = 1.56). Those non-price signals still point toward downside risk for duration-sensitive real estate. Price behavior, however, has recently diverged from our structural view: 3-day price_in excess (XLRE vs SPY) = +0.63%; etf window 3-day cumulative = +0.2%, max drawdown = 0.0% — tape has shown relative strength. Options IV is at a historical low (IV percentile 0%) and skew is more negative versus benchmark, indicating complacency and cheap downside protection but not a directional reversal. The triggering event (UK Q2 stronger growth → blueprint: rate_decision_hawkish_surprise) is a plausible global-hawkish narrative that would increase rate/duration pressure, but it is not direct evidence about RBA policy (historical_context = first_signal), so its causal link to Australian monetary conditions and to XLRE is weak. Therefore, while non-price evidence continues to favor short, the weakening of the systemic-liquidity argument (smaller 4-week net liquidity drawdown versus prior readings), plus recent relative price strength, reduce conviction. To materially strengthen the short call we would need direct, non-price evidence tied to RBA or US rates (e.g., realized AU CPI surprise higher, explicit RBA hawkish guidance, or a sustained rise in US Treasury yields). | +0.6% | 2026-08-19 |
| RBA Australia Rate Cycle rba_australia_macro_policy | KBE SPDR S&P Bank ETF | Contested | 0.30 | Maintain contested. The prior (2026-08-22) assessment was appropriately cautious: there is still no arc-specific non-price shock linking the RBA/Australia CPI to US regional bank valuations (KBE). Independent non-price evidence in this refresh does not change that. Quantitative facts: price_in (KBE vs XLF 2026-08-19→08-21) = -0.55%; etf_price_window (9d) cumulative = -1.89%, max drawdown = -4.26% — price shows modest weakness but per rules is only corroborative. Non-price signals are mixed: AUM net outflows since 2026-08-14 = -$41M (-2.43% AUM) while 30d flow = +4.60% AUM; short interest = 15,145,952 shares (days-to-cover 11.2, 7-period cumulative +265.2%) indicating crowded shorts and squeeze risk; option ATM IV percentile is elevated (IV_pctile 86%; call IV 36.57% / put IV 25.37%; skew -11.21pt) indicating higher uncertainty but skew is not a directional source here; Fed net liquidity 4w change = -$125B, a mild risk-reducing backdrop. Crucially, none of these non-price numbers form the arc-specific causal chain (RBA policy → AUD/global spreads → US banks → KBE). Therefore remain contested and await a clear arc-specific non-price trigger before taking a directional mid-term position. | -0.1% | 2026-08-19 |
| RBA Australia Rate Cycle rba_australia_macro_policy | AGG iShares Core U.S. Aggregate Bond ETF | Contested | 0.30 | Remain contested. The prior critical point — the trigger is not an Australia/RBA event and thus lacks the direct causal chain (domestic data → RBA decision → Australian rates/term premium → AGG) — still holds. Quantitative checks show mixed signals with no clear arc-specific direction: price_in excess (AGG vs SPY, 3d) = -0.05%, etf_price_window cumulative = -0.48%, max drawdown = -0.48%, latest close = 97.35 (2026-08-21), indicating only modest price movement without a decisive directional signal. Non-price metrics are mixed and broad-market in nature rather than arc-specific: AGG AUM $138.0B with net creation +$117M since 2026-08-17 (+0.08% AUM) and 30d +0.42% AUM (modest inflows); options show elevated protective demand (ATM IV historical percentile 92%; put IV 8.13% vs call IV 4.64%; put-call skew +3.49pt, ~+4.65pt vs SPY) — higher downside protection costs but skew alone is not a directional source; rates remain elevated at 10Y=4.69% with 2s10s=0.5 (term premium/long-end pressure); MOVE 73.4 (lower). Treasury auction demand (2-run avg bid-to-cover ≈ 2.67) is firm. None of these provide a direct, non-price quantitative link to an RBA policy inflection that would move this arc from contested to directional. Therefore keep contested (confidence 0.30). Note: Australia CPI is scheduled in ~2 days (2026-08-26); a material surprise there would be the appropriate arc-specific non-price trigger to reassess. | -0.1% | 2026-08-19 |
| BoE UK Gilt Cycle boe_uk_gilt_cycle | TIP iShares TIPS Bond ETF | Contested | 0.30 | Keep contested. The prior assessment (2026-08-22, contested 0.30) argued price and flows supported TIP but structural non-price rate signals (10Y real > breakeven) and lack of a clear BoE dovish trigger left direction ambiguous. That judgement still holds. Quantitative evidence shows mild support for TIP: price_in excess (15d) = +0.25%; etf_price_window (9d) cumulative = +0.22% with max drawdown = -0.36%; TIP 30-day net flows = +3.84% of AUM and since 2026-08-17 net inflows = +$161M (~+1.07% AUM). These flow/size metrics are non-price, quantitative supports. Offsetting this, FRED data show 10Y real ≈ 2.35% slightly above 10Y breakeven 2.34% (a mild headwind for TIPS), and today's trigger is a media forecast of UK growth slowing — narrative, not a quantified BoE policy signal. UK PMIs actually do not corroborate a slowdown (services 52.8 vs exp 51.8; manufacturing 51.5 vs exp 51.6). In sum: price and flows provide modest bullish market-behavior evidence, but rate-structure and lack of a BoE policy trigger keep the view mixed; remain contested with unchanged confidence. | -0.8% | 2026-08-18 |
| US-China Tech Decoupling us_china_tech_decoupling | UNG United States Natural Gas Fund | Short | 0.30 | Maintain a medium-term short tilt on UNG, but reduce conviction further. The primary non-price, arc-relevant drivers still point negative for long positions: 1) CFTC shows a net speculative short position (net = -110,630 contracts as of the 2026-08-11 report), and 2) the Henry Hub term structure remains in contango (front 2.775 vs 12m 3.198 → slope -13.227%), creating persistent roll/negative carry for long ETPs. Those form the causal chain relevant to this arc (contango → roll loss → ETF underperformance; COT net short → market positioning biased toward shorts). However, independent, non-price evidence in the latest snapshot weakens execution confidence: UNG has seen substantial net creations since 2026-08-03 (+$52M, +10.33% AUM; 30d +11.99% AUM), indicating crowded/flow-led long demand; ATM option IV is at the 12th percentile (cheap protection, lower squeeze cost) with put-call skew +1.37pt. Price/window metrics: arc-window cumulative UNG -5.30% (2026-05-08 → 2026-08-20), latest close 10.01 (2026-08-20), recent 5d +0.40%. Net-net: structural COT + contango continue to justify a short bias, but fund flows and compressed IV materially raise squeeze/rebound risk, so reduce confidence to 0.30 and favor smaller size and active protection/laddered execution. | -9.3% | 2026-08-15 |
| US CRE / Office Crisis Cycle us_cre_office_crisis_cycle | IYR iShares U.S. Real Estate ETF | Contested | 0.30 | Remain contested: the structural thesis (US office/CRE undergoing repricing via higher underwriting/insurance costs reducing NOI and higher financing costs compressing valuations) still holds, but current independent non-price signals do not provide a clear trigger for a 1–3 month directional trade. Quantified evidence: price_in excess = -1.42% (IYR cumulative +0.79% vs SPY +2.21%; 30-day cum 0.79%, max drawdown -4.37%) — relative weakness is present but not explosive; real-money flows since 2026-08-14 show net inflows of +$16M (+0.34% AUM) and 30-day net +0.54% AUM, indicating no broad forced selling; options show low absolute IV (call IV 13.82%, put IV 15.36%, IV historical percentile 0%) while put-call skew = +1.54pt (delta vs SPY skew = +2.70pt), implying concentrated downside hedging rather than fully priced systemic downside; Fed net liquidity has declined but not in an acute step (-$125B over 4w; total $5,792B); VIX is normal (16.01). Combined: the cre_office_crisis_repricing blueprint remains the organizing narrative, but the current non-price metrics (modest inflows, low IV, elevated skew, limited liquidity drain) argue against an immediate, high-conviction directional trade. A change would require a clear, arc-specific non-price shock (e.g., rapid CRE financing defaults, large insurance-premium shock, or bank balance-sheet retrenchment tied to CRE). | -1.4% | 2026-08-14 |
| RBA Australia Rate Cycle rba_australia_macro_policy | PFF iShares Preferred and Income Securities ETF | Contested | 0.30 | Remain contested. The prior assessment correctly flagged the absence of Australia-specific, non-price quantitative evidence (e.g., Australian CPI surprise, RBA minutes showing policy pivot, or Australia-specific COT/labor inflection) required to resolve the RBA arc. Today's independent snapshot does not change that structural view: price_in (3d PFF vs SPY) excess = +0.01% (PFF cumulative -0.42% vs SPY -0.43%, effectively neutral); etf_window 3d cumulative -0.42%, max drawdown -0.46%, latest close 30.46. PFF AUM = $13.1B with net creations/redemptions since 2026-08-14 = -$60M (~ -0.46% AUM), 30d flow -1.72%, 90d -1.95% (continued modest outflows — a fund-level liquidity/sentiment signal, not a direct RBA-policy signal). Options: call IV 12.35% / put IV 11.52% / skew -0.83pt (relative to SPY skew difference +0.33pt), IV percentile 40%. Credit: HY OAS 2.75%. Macro: U.S. curve steeper (2s10s = 0.5; 10Y = 4.69%) but not Australia-specific. No independent non-price evidence that ties through the RBA → Australian rates → PFF causal chain, so the arc remains contested. Watch upcoming Australian CPI (2026-08-26) as the likely next arc-specific trigger. | +0.0% | 2026-08-19 |
| US-China Tech Decoupling us_china_tech_decoupling | CQQQ Invesco China Technology ETF | Contested | 0.28 | Maintain 'contested'. The structural downside channel (Beijing regulatory/export-control or industrial policy → slower cross-border M&A / investment and valuation pressure → selective weakness in China tech) remains logically plausible, but there is no new non-price evidence that directly links a regulatory event to CQQQ’s path. Quantitatively: arc window price_in excess = −12.83% (CQQQ −8.22% vs FXI +4.61), etf_window cumulative −8.22%, max drawdown −9.04%, latest close 49.37 — prices have already reflected part of the downside. At the same time, independent non-price signals conflict: CQQQ AUM = $3.1B with 30-day cumulative inflow +6.14% AUM, today’s dollar turnover $62.8M vs 20-day baseline $39.8M (1.63x, flagged bullish_surge), short interest 3,246,901 shares (−31.1% vs prior), days-to-cover 2.66; options: call IV 40.82% / put IV 28.91% (skew −11.91pt), IV percentile 17% and 5-day IV −6.03pt. Price behavior supports prior downside thesis (partially realized), but flows/positioning indicate active inflows and short covering that counter a unilateral bearish call. Therefore the signal set is mixed and I keep the arc contested until clear, arc-specific non-price evidence appears (e.g., measurable M&A cancellations tied to regulatory action, explicit export-control rulings, or quantifiable financing freezes). | -9.7% | 2026-08-14 |
| China Rare Earth Export Restriction china_rare_earth_export_restriction | LIT Global X Lithium & Battery Tech ETF | Contested | 0.25 | Maintain contested. The prior assessment correctly highlighted the decisive constraint: absence of a non-price, lithium-specific quantitative link (e.g., COT, inventories, production, export volumes) that would map a policy event into lithium supply/demand and thus into LIT. Today's trigger — “China eases curbs on Nvidia H200 chips” — is a targeted regulatory easing for semiconductors/AI compute hardware and does not establish a causal transmission to lithium supply or demand (event → lithium inventories/production/export change → LIT). Quantitatively, price-derived signals remain weak: arc-window cumulative LIT -14.37% vs MXI +2.22%, LIT-MXI excess -16.59%, max drawdown -25.5%, latest close 76.61 (2026-08-21); these are consistent with historical export-control analogs (avg t+20 roughly -15% to -18%), but per rules price can only be a cross-check, not the sole directional driver. Non-price quantitative evidence available is limited to positioning/liquidity context: options crowding labeled call_crowded (put_call_vol_ratio ≈ 0.37), short interest 696,192 shares (-27.1%) with days-to-cover 3.07, AUM $1.6B and 30d flows -1.77% AUM. Those signals inform positioning and crowding risk but do not provide the required, arc-specific supply/demand causal link. The chip-export easing arguably reduces the probability of a broader, simultaneous export crackdown, but absent lithium-specific non-price metrics the event is insufficient to change the arc decision. Therefore remain contested with confidence 0.25. | -18.1% | 2026-08-21 |
| EU-US Tariff Dispute eu_us_tariff_dispute | XRT SPDR S&P Retail ETF | Contested | 0.25 | Maintain contested. The prior structural view still holds: event_opp #3801 (Europe ‘shrugs’) lowers the near-term probability of an EU‑US tariff shock, so the tariff_policy_shock channel alone does not generate a clear directional trade. Independent, non‑price quantitative signals remain bearish and have not been reversed: net real creations/redemptions since 2026-08-13 = -$164M (−44.50% of AUM); 30‑day cumulative flows = -19.60% AUM; FINRA short interest = 20,468,454 shares (+17.8%), days‑to‑cover = 4.17. Those funding/positioning signals exert downward pressure on XRT but were already present in the prior update. Price evidence is neutral-to-mildly negative: arc price_in excess = -0.09%, etf_price_window cumulative = -0.15%, max drawdown = -6.09%, latest close 87.71 (2026-08-21). Options IV is low (call IV 22.0% / put IV 20.12%, skew -1.88pt) — informative on crowding but not a standalone direction source. Net: event risk is muted (reduces a clear bullish tariff-repricing case) while flows/positioning sustain downside friction → keep contested. | -2.8% | 2026-08-19 |
| US EV Subsidy & Buildout Cycle us_ev_subsidy_buildout | LIT Global X Lithium & Battery Tech ETF | Short | 0.24 | Maintain prior short stance (low conviction). The causal chain remains: Samsung SDI's acquisition of GM's 49.99% stake in SynergyCells and refocusing the New Carlisle plant toward energy storage/non-passenger applications implies greater vertical integration and a shift of some battery output away from passenger EVs—reducing near-term external battery and upstream lithium demand (event_opp #3843, 2026-08-16). There are no new corporate/policy events or asset-specific non-price drivers (COT/IV/predictive markets) to overturn the thesis. Independent quantitative signals are mixed: 30-day fund flows remain net outflows at -1.77% AUM (as of 2026-08-19), which is bearish; but FINRA short interest has fallen to 696,192 shares (settlement 2026-07-31, -27.1% vs prior) with days-to-cover=3.07, indicating short covering that reduces downside crowding. Price-wise, the ETF 5-day window shows +0.49% cumulative return and max drawdown -2.91%, while arc-window LIT vs MXI excess is -4.00% (2026-08-17→08-21), i.e., LIT remains weak relative to the broader materials index. Because no new independent non-price evidence strengthens or flips the causal chain, I keep the short position but keep conviction low due to partial price recovery and reduced short crowding. Historical-analogue strength remains low (first_signal); the arc is still early so inertia half-life likely not yet passed, but decay risk exists and should be monitored. | +1.4% | 2026-08-16 |
| Clean Energy Transition clean_energy_transition | TAN | Contested | 0.24 | Maintain contested: the prior view (2026-08-23) correctly judged the Sunshine policy as a narrative/regulatory/reshoring signal without a quantifiable, non-price causal chain linking the narrative to concrete subsidy/orders/demand that would justify a directional trade in TAN. In this daily refresh (2026-08-24 07:00 ET) independent non-price evidence remains insufficient to close that chain: options IV chain quality is sparse (TAN IV missing), FINRA short interest = 1,357,912 shares (−1.9% vs prior), days-to-cover = 1.13 (settlement lag; ambiguous squeeze signal; 7-period cumulative −7.0% shows recent short reduction but no clear one-sided pressure); macro/market backdrop is mildly favorable (market breadth = 73.8% > 50-day MA; VIX term ratio = 0.859; HY OAS = 2.75%; Fed net liquidity = $5,792B, 4-week −$125B). These quantitative datapoints are background or crowding/risk indicators and do not provide the required, arc-specific evidence (e.g., concrete subsidies, multi-year procurement contracts, industry-level supply constraints) that would causally link the event to TAN outperformance. Price-derived metrics (price_in / excess_return / etf_window cum/drawdown) are unavailable (T-1 close missing) and therefore cannot confirm/negate the thesis. Conclusion: remain contested. Actionable directional conviction requires new, non-price, arc-specific data such as announced federal subsidies, long-term off-take agreements, or clear option-flow/IV shifts indicating one-sided capital commitment. | — | 2026-08-17 |
| Trump 2026 Tariff Policy trump_tariff_2026_policy | EWT iShares MSCI Taiwan ETF | Contested | 0.22 | Maintain contested. The prior assessment concluded there was no independent non-price evidence to connect 'Trump 2026 tariff/export-control' to a quantifiable hit to Taiwan equities/EWT fundamentals, so no clear directional call. No new policy actions, prediction-market moves, or CFTC-style positioning changes have appeared to close that causal gap. Instead, independent metrics point to greater crowding: net creations +$104M (+0.92% AUM) since 2026-08-14 (30d +5.33% AUM), cheap option protection (IV historical percentile 8%) and EWT skew -3.70pt vs EEM -2.66pt (delta -1.04pt), and rapid short accumulation (FINRA short shares 8,062,620, +37.6%, days-to-cover 1.19). Price-derived evidence shows 6d price_in excess = +0.84% and 2-day ETF window cumulative +0.22% (EWT‑EEM excess -0.53%), which do not resolve the dispute. Therefore, lacking new non-price causal evidence, but seeing stronger crowding/low-IV signals that raise tail risk, I keep the arc contested and reduce conviction (weaken). | -0.2% | 2026-08-20 |
| Clean Energy Transition clean_energy_transition | SLX VanEck Steel ETF | Contested | 0.20 | Remain contested. The prior assessment correctly flagged the absence of a non-price, arc-specific causal chain linking the event (industrial metals / supply squeeze) to SLX. Current quant facts show price_in excess = -2.07% (SLX window cumulative = -1.02%, max drawdown = -3.23%), i.e. prices have not realized a one-sided bullish outcome. Non-price metrics exist (30d AUM flows +1.89% AUM; FINRA short shares +143.6% with days-to-cover = 2.28; 4-week Fed net liquidity -$125B; VIX term ratio = 0.859) but are mixed, general market/flow signals and crucially lack the required “on‑point” supply/inventory/COT/production data to close the causal chain (event → supply change → metal prices → SLX). Therefore we cannot take a directional stance for the 1–3 month arc; remain contested with low conviction. | -0.1% | 2026-08-17 |
| Clean Energy Transition clean_energy_transition | KARS | Contested | 0.20 | Keep contested: the narrative remains fiscal/regulatory upside (fiscal_clean_energy_subsidy + regulatory_action_chinese_tech) but there is still no quantified, arc-specific causal chain that would convert that narrative into a clear directional bet on KARS. Structured checks: price_in_excess_return = unavailable; etf_window_cum/drawdown = unavailable; historical_analog_strength = low; options IV chain sparse/unavailable — so price/vol structure cannot be used alone to set direction. The most salient independent non-price signal this refresh is the FINRA short build (short shares = 42,100, +1109.1% vs prior, 7-period cumulative +52.7%, days-to-cover = 3.5), which raises short‑crowding / squeeze risk — discouraging a naked short but not providing a quant confirmation of a long. Macro/market backdrop (market_width = 73.8% > 50%; VIX term ratio = 0.859 contango; Fed net liquidity = $5,792B 4w -125B; EU/DE manufacturing PMI surprise) is mildly supportive to risk assets but is not arc-specific. No new arc-specific non-price quantitative evidence (e.g., formal subsidy amounts, signed procurement orders, or measurable policy effective dates) appeared to flip or strengthen the prior view. Maintain contested, confidence 0.20. | — | 2026-08-17 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | MCHI iShares MSCI China ETF | Contested | 0.20 | Maintain contested. The prior assessment correctly concluded there is no single non-price quantitative signal that decisively supports a long or short on MCHI; I agree. Price-derived evidence has partly realized: price_in 5d excess = -4.47% (MCHI vs EEM), 9-day window cumulative = +0.07%, max drawdown = -2.16%, close = 55.66 — indicating modest underperformance but no large structural breakdown. Independent non-price quantitative indicators remain mixed and non-directional: VXEEM = 27.05 (60d z = -2.06, EM volatility unusually low); MCHI 30-day net flows +0.01% AUM (AUM $6.3B); daily dollar volume $58.1M vs 20-day baseline $131.5M (0.44x, low activity); option chain quality sparse (put IV = None, call IV = 19.29%); FINRA short interest = 15,353,069 (7-period cum +19.1%), days-to-cover = 5.94 (shorts crowded); prediction market yes_prob(Fed hike in 2026) = 0.47 (neutral); Fed net liquidity (4-week) = $5,792B (4w -125B). These signals offset each other (low vol vs marginal liquidity withdrawal; short accumulation vs potential squeeze). No independent non-price evidence in the fresh snapshot strengthens a directional view, so I keep contested and modestly reduce confidence from 0.22 to 0.20 because price_in has partially played out and the option chain is sparse (lower information content). | +0.7% | 2026-08-11 |
| Korea / US Geopolitics korea_us_geopolitics | BNDX Vanguard Total International Bond ETF | Contested | 0.20 | Hold contested (cautious neutral). There is no new non-price quantitative evidence that meaningfully strengthens the causal chain (Korean Peninsula → safe-haven duration inflows). Market-implied Fed funds at 3.735% implies some easing expectations (supportive for duration), but long-end signals oppose—10Y=4.69% and term premium=0.8393% exert pressure on duration. Treasury auction bid-to-cover ≈2.67 (recent runs) points to demand support, while MOVE=73.4 indicates no acute spike in rate volatility. Price action partially reflects the mixed forces: BNDX price_window cumulative excess = -0.75%, 52-day cumulative -0.75%, max drawdown -1.92%, recent 5d -0.15% (price does not validate a single-sided safe-haven move). The current event (Carson Block / passive investing commentary) is unrelated to Korea-US geopolitical escalation and therefore neutral/noise. Prediction-market invasion probability is low (0.03). With non-price indicators conflicted and crowding signals present (BNDX short-interest +53.8% with days-to-cover=1), maintain a contested stance and avoid committing a larger directional exposure. | -2.3% | 2026-08-17 |
| US EV Subsidy & Buildout Cycle us_ev_subsidy_buildout | URA Global X Uranium ETF | Contested | 0.20 | Prior (2026-08-22) set the arc to contested: the Samsung SDI acquisition was a corporate-level battery event lacking a uranium/nuclear-specific non-price causal link (e.g., uranium inventories, producer output/capacity changes, regulatory approvals or policy shifts). This refresh tests that conclusion. Price action: URA 5-day window cumulative +1.81% but price_in excess vs PICK = -2.39%, 5-day max drawdown -3.6%, latest close 46.07 — a mixed tape move that does not establish a uranium fundamental story. Non-price signals include ETF net creation +$57M (+0.90% AUM since 2026-08-05), 30d +1.12% AUM, ATM option IV at historical percentile 0% (cheap protection), FINRA short interest +11.8% to 3,420,445 shares (days-to-cover=1.1), and today’s volume $191.2M (1.38x 20-day baseline). Those are funding/flow/crowding indicators and risk-management inputs but they are not arc-specific causal evidence linking the corporate battery event to uranium supply/demand or policy (no uranium inventories, no Cameco/Cameco-like production change, no NRC licensing or subsidy action). Therefore keep contested: do not take a directional position nor raise conviction. The low IV and rising short interest are crowding/fragility signals for execution risk, not directional proof for the arc thesis. | +2.7% | 2026-08-16 |
| RBA Australia Rate Cycle rba_australia_macro_policy | IYR iShares U.S. Real Estate ETF | Contested | 0.18 | Remain contested. The prior assessment correctly treated the trigger as non-Australia-specific (FTSE / UK mining story) and therefore not a direct empirical signal for the RBA path. Price cross-check since the prior note shows price_in excess = +0.38% (IYR vs SPY, 3d), while IYR’s own 3‑day cumulative return is only -0.06% with max drawdown -0.06%, indicating no clear pricing-in of an RBA pivot. Independent non-price signals in this refresh slightly weaken the bullish case: IYR net creations/redemptions since 2026-08-17 = -$110M (−2.41% AUM), 30d flow = -0.94% AUM; option metrics show put-call skew = +1.54pt (vs SPY skew −1.16pt → +2.70pt delta) and IV percentile = 0% (9‑day), with 5d IV −0.84pt and skew +0.98pt. These fund-flow and options signals point to modest capital withdrawal and a higher price of downside protection for IYR, but they do not form a causal, Australia-linked chain (no RBA decision, no Australia CPI surprise, no prediction‑market shift). So direction remains contested, but confidence is reduced from 0.22 to 0.18. Watch the Australia CPI (2026‑08‑26) — a true arc-relevant non-price surprise there would be the proper trigger to change direction. | +0.4% | 2026-08-19 |
| RBA Australia Rate Cycle rba_australia_macro_policy | VNQ Vanguard Real Estate ETF | Contested | 0.18 | The prior assessment (2026-08-22: contested, confidence=0.18) framed the chain: hotter Australian macro → RBA more cautious on cuts / relatively hawkish → higher Aussie yields → discount-rate pressure on rate-sensitive REITs (VNQ) supporting a short leg. That prior also noted the scarcity of additional AU non-price evidence beyond the earlier unemployment surprise, and that market/derivative signals were not aligned with a short — hence contested. Today's refresh keeps that framing: 1) No new Australia-specific non-price data (no CPI/RBA communication) appeared to strengthen the hawkish-RBA transmission; 2) Market pricing and derivatives do not confirm short exposure — quantified: price_in excess (3d VNQ vs SPY) = +0.32% (VNQ slightly outperformed), etf_price_window 3d cumulative = -0.11%, max drawdown = -0.11%, latest close = 98.5; options: call IV = 15.75% / put IV = 12.16% (put-call skew = -3.59pt; skew diff vs SPY ≈ -2.43pt), IV percentile = 8% (cheap protection); VNQ short interest = 5,506,617 (-9.0% vs prior settlement), days-to-cover = 1.9; volume proxy $285.2M vs 20d $288.7M (0.99x, neutral). In sum: the sole direct non-price cue (AU unemployment surprise) remains but is not reinforced; price/IV/flow signals do not back the short. Therefore the arc remains contested with low conviction (0.18). A clear AU non-price trigger (e.g., CPI/RBA commentary) would be required to move this assessment decisively. | +0.3% | 2026-08-19 |
| China LFP Battery Dominance china_lfp_battery_dominance | LIT Global X Lithium & Battery Tech ETF | Contested | 0.15 | Remain contested. The prior assessment correctly highlighted the core tension: structural Chinese policy/subsidy and export-control dynamics imply medium-term downside for certain lithium-relevant chains, but market price and flows have been resilient and there is still no arc-specific non-price quantitative evidence to take a one-sided stance. Quantitatively, price-in has materially occurred and conflicts with historical analogs: etf window (2026-07-22→2026-08-21) cumulative +11.03%, max drawdown −3.3%, latest close 76.61 (2026-08-21), arc excess LIT−MXI = +1.39%; by contrast historical analogs for export-control events showed avg t+20 return ~ −15.0%, so observed price action does not align with those negatives. Non-price evidence remains insufficient to pick a direction: options IV is missing (unusable), there are no lithium-specific COT/inventory/production or concrete policy implementation metrics; available non-price signals are contextual only (EU manufacturing PMI = 52.8 σ=1.2; DE manufacturing PMI = 54.1 σ=2.0; Fed net liquidity 4-week = $5,792B, −$125B; LIT 30d AUM flow −1.77%). Conclusion: keep contested, but reduce confidence because price-in has already run and arc-specific non-price signals remain absent. Await arc-specific quantitative triggers (COT/inventory/production shifts, policy implementation metrics, or usable IV/option signals) before taking a directional position. | +8.6% | 2026-08-17 |
| US EV Subsidy & Buildout Cycle us_ev_subsidy_buildout | TAN | Contested | 0.15 | Samsung SDI’s purchase of GM’s JV stake and the repurposing of the Indiana plant toward energy storage/high-tech applications creates offsetting transmission channels: weaker EV end-market demand (negative for EV-panel/battery exposure) versus potential positive reallocation to storage/adjacent supply-chain winners. Since the prior update there is no independent non-price quantitative evidence that singles out a direction: systemic liquidity has declined (WALCL−TGA−RRP = $5,792B, 4w Δ = −$125B as of 2026-08-19), which raises risk premia but is not a direct, arc-specific signal for TAN; TAN short interest has not meaningfully expanded (FINRA short shares = 1,357,912; −1.9% period change; days-to-cover = 1.13), showing crowding but no large squeeze/reversal signal; options IV chain is missing (IV chain_quality = sparse) and cannot be used as an independent directional input; price-derived indicators (price_in / excess_return / etf_window cum/drawdown) are unavailable, so we cannot tell whether the market has already priced this. The historical analog remains a first_signal with low strength and arc-coverage heat has fallen. Given offsetting causal channels and no new independent non-price evidence to pick a side, maintaining a contested stance at low confidence is appropriate. | — | 2026-08-16 |
| Clean Energy Transition clean_energy_transition | COPX Global X Copper Miners ETF | Contested | 0.12 | Prior assessment (contested, confidence 0.20) correctly noted the structural chain (clean-energy → higher copper demand → potential supply tightness) and that CFTC COT provided a relevant non-price bullish signal, but that ETF outflows and rising short interest offset a one-sided bullish view; price had already partially realized the move. Self-critique: treating COT as heavily decisive was risky — that remains valid. Today's update: no new independent non-price confirmation for a bullish flip. CFTC net-long remains elevated but slightly lower (+78,648 contracts, 97th percentile as of 2026-08-18), still indicating crowded long positioning rather than fresh conviction. ETF funding flows remain net outflows (since 2026-08-10: -$139M, -1.60% AUM; 30d -0.96% AUM). FINRA short interest increased (4,691,947 shares, +5.5%, days-to-cover 1.84), implying either persistent hedging/selling pressure or squeeze risk. Price has partially realized the thesis (price_in excess +4.10%; COPX 5d +8.30% vs PICK +4.20%; etf_window cum +8.3% / max drawdown -2.03%) with a volume surge (2.08x baseline). Options IV is neutral (call IV 48.79% / put IV 47.63%, IV percentile 50%), skew is not a reliable directional source. Conclusion: the structural bullish chain and COT remain present but not newly strengthened by independent non-price data; given partial price realization and persistent mixed non-price flows, retain contested but lower confidence (weaken). Estimated arc age ~7 days vs plausible half-life ~14 days → approaching decay. | +9.2% | 2026-08-17 |
| Clean Energy Transition clean_energy_transition | URA Global X Uranium ETF | Contested | 0.10 | Remain contested. The prior rationale stands: there are no 'look-alike' non-price uranium/nuclear supply–demand metrics (e.g., reactor permits/completions, uranium production/inventories, COT-style positioning) that create a credible event→uranium demand→URA causal chain, so we cannot take a mid-term directional position. Quant checks: 5‑day ETF window shows URA cumulative +1.81% with max drawdown −3.6%; matched peer PICK +4.20%, giving URA‑PICK excess = −2.39% (partial price‑in of relative weakness). Non‑price signals are conflicting or non‑specific to uranium: AUM net creations since 2026‑08‑05 +$57M (+0.92% AUM; 30d +1.15%) suggests modest inflows, while FINRA short interest at 3,420,445 shares (+11.8% vs prior) and days‑to‑cover 1.1 indicate rapid short accumulation/short‑side crowding. ATM option IV sits at historical percentile 0% (call 44.68%, put 42.38%) implying cheap protection/complacency but not a directional read alone. Dollar volume $191.2M vs 20‑day baseline $128.3M (1.38x) is elevated but only contextual. Event is a first_signal -> no strong historical analog. Conclusion: stay contested, slightly lower conviction because part of the thesis has been price‑realized and non‑price evidence remains non‑conclusive; watch upcoming macro (US core PCE/GDP) for risk management. | +2.7% | 2026-08-17 |
| Korea / US Geopolitics korea_us_geopolitics | EUAD Select STOXX Europe Aerospace & Defense ETF | Contested | 0.10 | Remain contested. Structurally the Korea‑US geopolitics → European defence premium narrative is intact, but independent non‑price quantitative signals do not support a renewed, material escalation that would push EUAD significantly higher in the near term. Price has already partly priced this: arc window cumulative EUAD +13.15% with a +9.92% excess vs ITA, window max drawdown −9.18%, latest close 46.12 — indicating substantial price_in. Non‑price evidence tilts neutral/negative for an escalation‑driven long: prediction market probability of North Korean invasion = 0.03 (Polymarket, $440k vol) directly lowers the likelihood of a major trigger; EUAD AUM shows net outflows (30d −1.49%, 90d −2.82%), weakening fund flow support; option IV is at a low percentile (0% vs own 9‑day history) with 5‑day IV −2.65pt and skew −2.12pt (short‑term fear receding). Relative skew vs ITA is +2.54pt but skew alone is not a direction source. Macro PMI prints in Europe are stronger than expected (EU manuf 52.8, DE manuf 54.1), consistent with higher risk appetite and less safe‑haven bid. No new non‑price evidence justifies flipping to long/short; therefore remain contested and lower conviction to 0.10. | +12.2% | 2026-08-17 |
| RBA Australia Rate Cycle rba_australia_macro_policy | UUP Invesco DB US Dollar Index Bullish Fund | Contested | 0.09 | Maintain contested (no unilateral position). The prior judgement was correct: this arc requires an Australia-specific non-price trigger (RBA action or Australia CPI surprise) to justify a directional USD trade. No such arc-internal trigger has appeared; the next relevant scheduled release is AUS CPI on 2026-08-26. Quant facts and price-derived context: price_in (arc window excess) = +1.45%, etf_window cumulative +1.45% / max drawdown -2.52% (75-day window). Non-price snapshots do not support taking a clear directional USD stance: UUP AUM = $0.3B with 30-day net redemptions -32.38% AUM (5 data days), 90-day -32.83% (6 data days) indicating capital outflow; FINRA short interest +22% (settlement 2026-07-31) with days-to-cover = 3.22 (shorts accumulating). Options: ATM IV low (call IV 7.76%, put IV 6.35%, IV percentile 47%, skew -1.42pt) — descriptive only. Macro/liquidity backdrop: US 2Y=4.19%, 10Y=4.69%, broad USD =118.9028 (20d -1.4%), VIX term structure calm (VIX/VIX3M=0.859). In sum: absent arc-specific RBA/AU non-price evidence and given fund outflows plus modest price_in, keep contested but with reduced confidence. | -4.3% | 2026-08-19 |
| NATO / EU Defense Spending Cycle nato_eu_defense_spending | XAR SPDR S&P Aerospace & Defense ETF | Contested | 0.08 | Carry forward the prior view (2026-08-23): continue to track XAR (NATO/EU defense spending arc) but make no directional bet (contested). Recap: the last assessment noted absence of arc-specific, quantitative policy/budget/contract evidence to justify a one-sided long, and that price had already declined, partially discounting any prior premium. Today's check finds no material change. Non-price quantitative signals remain mixed: passive creation shows small net inflows since 2026-08-12 (+$29M, +0.47% AUM; 30d +1.76% AUM), indicating limited passive allocation, while institutional shorts have accumulated rapidly (FINRA short = 295,329 shares, +45.2%; days-to-cover = 1.26), a non-price sign of bearish positioning and crowding risk. Price has already realized downside: price_in_excess = -6.61%; etf_window cumulative = -7.51%; max_drawdown = -7.83%. The recent event (Macron speech, event_opp #4170) is narrative/supportive but lacks quantified budgetary or contract data that would constitute the required 'on-the-nose' non-price driver for a committed long per the blueprint. Prediction markets for NATO×Russia remain low (polymarket 8/31 yes_prob = 0.04; 12/31 = 0.24). Historical-analog strength is weak (first_signal). Synthesis: no new, quantifiable, arc-specific evidence to change direction; price has continued to discount risk and positioning is mixed, so maintain a contested stance and modestly lower conviction (0.10 → 0.08). | -6.6% | 2026-08-15 |
| NATO / EU Defense Spending Cycle nato_eu_defense_spending | EWG iShares MSCI Germany ETF | Contested | 0.08 | Remain contested. The prior assessment correctly identified the absence of 'on-the-nose' non-price triggers (e.g., official EU/NATO budget uplifts, signed large procurement contracts, or verifiable order flow in constituent companies) needed to form a causal chain (event → budgets/contracts → constituent profits → ETF). No such on-the-nose evidence has appeared since the prior note. The observable changes are mainly price and positioning being partially priced: arc price_in excess = +0.74% (EWG +0.55% vs EFA -0.19%), etf_window 5d cumulative = +0.55% with max drawdown -0.64%. Non-price quantitative signals exist at the positioning level (30d creation/redemption +4.55% AUM, 90d +5.99% AUM; short interest 3,491,402 shares, -23.9%, days-to-cover = 2.95) but these reflect flows/positioning and do not substitute for causal evidence of increased defense spending. Prediction market for NATO×Russia by Aug31 is low (yes_prob = 0.04), reducing the near-term probability of conflict-driven procurement. Options show put-call skew = -1.42pt (EFA-relative +1.50pt) and IV percentile = 26% — useful as risk/positioning context only. Conclusion: No new, verifiable defense-budget/procurement data to change the prior view; price and flows have partially priced the narrative and reduce marginal informational value. Keep contested with low confidence (0.08) and await explicit on-the-nose non-price triggers before taking a directional ETF position. | +1.4% | 2026-08-15 |
| US-China Tech Decoupling us_china_tech_decoupling | DBA Invesco DB Agriculture Fund | Contested | 0.08 | Remain contested. The prior assessment (2026-08-23, contested, confidence=0.08) stands under today's refresh: 1) Price has partly realized the theme — price_in_excess = +6.27% (etf_window cumulative +6.27%, max run-up 6.49%, max drawdown −2.87%, last close 28.32 for the 2026-06-22→2026-08-21 window), so do not upsize directional exposure without new structural evidence; 2) There is no non-price, supply-side data (USDA/COT/inventories/exports) that links the sovereign/sanctions event to agricultural supply and DBA; 3) Available non-price quantitative signals do not support a clear directional read: short interest 291,535 shares (−52.4%) with days-to-cover = 1 shows shorts have shrunk; options IV is low (call IV 17.87%, put IV 16.50%, IV percentile 32%), implying the market is not pricing a large tail risk; dollar volume $30.2M vs 20d $22.2M (1.34x) is system-tagged neutral. Given absence of a causal non-price chain and the realized price excess, keep contested at confidence 0.08. Note: near-term macro prints (US core PCE, GDP on 2026-08-26) increase near-term event risk — avoid adding directional exposure pre-event. | +3.4% | 2026-08-15 |
| US-Iran 2026 War us_iran_2026_war | URA Global X Uranium ETF | Contested | 0.07 | Remain contested. The prior assessment correctly flagged lack of uranium-specific non-price evidence and therefore refused to take a directional view; that core gap remains. Quantitative recap: arc window (2026-06-16→2026-08-21) URA cumulative -3.76%, URA–PICK excess -3.91%, max drawdown -21.62%, latest close 46.07 (2026-08-21; 1d +5.09% is a tape outlier and price_in shows no sustained excess). Fresh non-price signals are mixed and offsetting: fund flows show small net inflows (AUM $6.1B; net +$57M since 2026-08-05 ≈ +0.92% AUM; 30d +1.15% AUM) and FINRA short interest has risen to 3,420,445 shares (+11.8%, days-to-cover=1.1) implying crowding/short accumulation; but prediction markets still price a high probability of de‑escalation (Polymarket: ceasefire through 8/31 yes_prob=0.92) and options show cheap protection (ATM IV percentile 0%, 5d IV -3.65pt, put-call skew -2.29pt), indicating market complacency on a sustained geopolitical risk premium. Per rules for commodity/energy/flight-to-safety ETFs, we need direct uranium supply-side non-price data (COT, inventories, production/ban orders) to take a committed directional stance—none exist in this refresh. Therefore we keep a contested posture and reduce confidence (0.10 → 0.07). A directional change would require new, uranium-specific quantitative evidence (e.g., inventory shocks, COT shifts, or explicit sanctions/ban with quantified impact). | -5.8% | 2026-08-16 |
| China LFP Battery Dominance china_lfp_battery_dominance | KWEB KraneShares CSI China Internet ETF | Contested | 0.06 | Keep a contested stance (no directional position). The prior assessment was correct: there is still no 'on‑the‑nose' non‑price quantitative evidence that converts a structural LFP battery thesis into a clear directional trade for KWEB (we need an event → asset‑specific metric → company fundamentals → ETF causal chain). No new on‑the‑nose events have appeared since the last update. Price has already absorbed part of the expected move: price_in excess = -3.32% (KWEB +0.83% vs FXI +4.15%, 2026-07-22→2026-08-21), 9d etf_price_window cumulative = -6.51%, max drawdown = -7.80% — implying realized price movement and reduced upside from the arc. Independent non‑price signals remain mixed/neutral: option IV percentile = 9% (call IV 29.49% / put IV 27.05% / skew -2.44pt vs FXI -2.34pt), VXEEM = 27.05 (z = -2.06, low vol regime), FINRA short interest = 41,541,170 (7‑period cumulative +16.4%, days‑to‑cover = 1.93) — short accumulation but still squeeze risk. Because the only clear changes are price realization and time decay, and no new on‑the‑nose non‑price evidence has appeared, I lower conviction (0.10→0.06) rather than strengthen or flip the prior view. | -1.6% | 2026-08-17 |
| BoJ Yen Normalization boj_yen_normalization | MCHI iShares MSCI China ETF | Contested | 0.06 | Remain contested (no directional call). I acknowledge and largely agree with the prior assessment: it flagged the absence of arc-specific, non-price quantitative evidence — and that absence persists in this refresh. Quantitative facts: price_in (2-day MCHI vs EEM excess) = +0.58% (MCHI +1.19% vs EEM +0.61%); etf_price_window (9d) cumulative = +0.07% with max drawdown -2.16%; arc window (2026-08-11→08-21) MCHI +0.07% / EEM +2.58% → MCHI-EEM excess = -2.51%, indicating no sustained outperformance consistent with a BoJ/JPY normalization story. Non-price snapshots exist but are not arc-specific: 30d AUM flows +0.01% (neutral), dollar volume $58.1M vs 20d baseline $131.5M (0.44x, low turnover), VXEEM = 27.05 (60d z = -2.06, low vol regime), FINRA short interest 15,353,069 shares, days-to-cover 5.94 (shorts accumulated +19.1%, crowding context). Option IV is sparse (call IV 19.29%, put IV missing) and unsuitable as a directional anchor. Crucially, there is no BoJ/JPY-specific non-price driver (policy statement, JPY COT flip, prediction-market probability move) to link BoJ normalization → JPY → capital flows → MCHI. Historical analog remains 'first_signal' with low consistency. Therefore keep contested with low confidence (0.06). | +0.7% | 2026-08-11 |
| US-China Tech Decoupling us_china_tech_decoupling | WOOD iShares Global Timber & Forestry ETF | Contested | 0.06 | The prior (2026-08-23) view was contested because there was no arc-specific, non-price quantitative trigger and price had already partially priced in the narrative. Review of fresh data finds no new evidence that links event_opp #4082 (Venezuela sovereign/restructuring/humanitarian) into the US-China tech-decoupling causal chain. Quantitatively: price_in_excess = +1.54% (WOOD +8.03% vs MXI +6.50%), window max drawdown = -3.66%, latest close = 72.49 — prices have already reflected part of the story. Non-price snapshots do not provide arc-specific directional support: AUM net inflow +$4M (since 2026-08-17, +1.58% AUM; 30d cumulative +1.58%), and dollar ADV $1.4M vs 20-day baseline $1.3M (1.11x) point to fund flows/participation (crowding) rather than causal drivers for the arc; FINRA short interest 38,427 shares (+82%) with days-to-cover = 2.74 signals rapid short accumulation and squeeze risk, not a clean directional trigger. Other non-price indicators (prediction market China→Taiwan invade yes_prob=0.04; China-Philippines clash yes_prob=0.33; VIX term ratio = 0.859; Fed net liquidity $5,792B, 4-week -$125B) give macro/risk context but do not establish the required causal link. Conclusion: prior contested view remains appropriate — no new arc-specific non-price evidence to go long/short; price has partly realized the theme and the arc window (≈60 days) exceeds a typical half-life, so keep contested with low confidence (0.06). | +5.2% | 2026-08-15 |
| RBA Australia Rate Cycle rba_australia_macro_policy | XLF Financial Select Sector SPDR Fund | Contested | 0.06 | Remain contested: the prior assessment correctly flagged the lack of an arc-specific, non-price causal chain linking RBA/Australia policy moves to XLF. Since the last update there is no new non-price signal that ties Australian policy or AUD spreads directly to XLF, so we cannot move to a directional long/short. The reason for reducing conviction is independent, non-arc-specific funding and mid-term price evidence: since 2026-08-13 XLF has seen actual creation/redemption net outflows of -$1,728M (-3.07% of AUM) and 30-day cumulative flows -1.71% AUM—an independent cash withdrawal signal—while the arc etf_price_window (9d) shows cumulative -1.49% and max drawdown -2.25%, indicating mid-term price weakness. price_in_excess (3d) = +0.43% shows short-term tape divergence but is price-derived and cannot alone determine direction. Therefore, because fund outflows and mid-term drawdown strengthen the evidence that the market is withdrawing from XLF but do not close the arc-specific causal chain to RBA, confidence is lowered from 0.10 to 0.06 (weaken) and the stance stays contested. Historical-analog support is weak (first_signal). | +0.4% | 2026-08-19 |
| Africa Military Conflict africa_military_conflict | URA Global X Uranium ETF | Contested | 0.04 | Remain contested. The prior assessment correctly flagged the absence of a direct, non-price supply link (e.g., mine shutdowns, export bans, official uranium production cuts or COT/inventory indicators) that would validate the chain “Africa military conflict → uranium supply hit → URA outperformance.” Since the prior note, price data do not introduce structural transmission evidence: price_in excess = -0.15% (URA +12.64% vs PICK +12.79%), etf_window cumulative return 12.64%, max drawdown -8.78%, last close 46.07 (2026-08-21). Independent non-price signals also do not provide a supply shock signal. Fresh snapshot shows AUM = $6.1B and net creations +$57M (+0.92% AUM) since 2026-08-05; 30‑day flows +1.15% AUM. Options: call IV 44.68% / put IV 42.38% with IV historical percentile = 0% and 5‑day IV change -3.65pt (cheap protection, falling IV). Short interest rose to 3,420,445 shares (+11.8%), days‑to‑cover = 1.1; intraday dollar volume $191.2M = 1.38x 20‑day baseline. These data point to positioning/ crowding and cheap hedging rather than a verifiable supply disruption that would justify a directional view. Therefore keep the arc contested, and reduce conviction from 0.06 → 0.04 to reflect partial price realization (etf_window +12.64%) and heightened crowding/structure risk. Historical analogs remain weak (historical_context = first_signal) and the current event (gold/sanctions in Sudan) does not provide the required “对口” causal link to uranium markets. | +11.2% | 2026-08-12 |
| China LFP Battery Dominance china_lfp_battery_dominance | CQQQ Invesco China Technology ETF | Contested | 0.03 | Maintain contested. The prior assessment was correct: there remains no arc-specific non-price quantitative evidence (e.g., explicit LFP subsidy amounts, system-level capacity build schedules, or quantified industry supply/demand gaps) to support a directional trade in CQQQ. Quantified facts since first signal: price_in excess = -8.20% (CQQQ -4.04% vs FXI +4.15%), arc window cumulative -4.04%, window max drawdown -7.82%, latest close 49.37 (2026-08-21), and the arc age ≈31 days which exceeds the approximate half-life (~30 days), so conviction should decay. Fresh non-price datapoints (30d AUM cumulative inflow +6.14% AUM; today’s dollar volume $62.8M vs 20d baseline $39.8M = 1.63x, flagged bullish_surge; option IV percentile 17% with call IV 40.82% vs put IV 28.91% → skew -11.91pt; short shares 3,246,901 (-31.1%), days-to-cover=2.66; polymarket invasion yes_prob=0.04; risk backdrop: VIX term ratio 0.859, Fed net liquidity $5,792B (4w -125B)) describe market positioning/liquidity but do not provide a causal transmission from an LFP-policy/production shock to CQQQ returns. Therefore these are context/participation signals, not arc-confirming signals. Outcome: remain contested; lower confidence from 0.05 to 0.03 because of price realization and time decay. | -6.5% | 2026-08-17 |
| Clean Energy Transition clean_energy_transition | LIT Global X Lithium & Battery Tech ETF | Contested | 0.02 | Remain contested (no directional bias). The prior assessment correctly identified the weak causal transmission from the “Sunshine policy” story to lithium fundamentals: there remains no directly relevant non-price evidence (lithium COT, mine production/inventory shock, or options IV spike) to justify a directional trade. Since the last refresh there are no new independent pro-directional non-price signals; instead, there are slight negative non-price indicators and the market has partially priced the uncertainty. Quantitative anchors: price_in excess (LIT vs MXI, 2026-08-17→08-21) = -4.00% (LIT +0.49% vs MXI +4.49%); etf_price_window(5d) cumulative = +0.49%, max drawdown = -2.91% (close 76.61, 2026-08-21); LIT 30d cumulative flows = -1.77% AUM (AUM $1.6B, as of 2026-08-20); Fed net liquidity (4w) = -$125B (as of 2026-08-19). Missing: lithium COT and options IV (data sparse), no mine-level production/inventory metrics and no predictive-market signal that would close the causal chain. Price moves are treated as confirmation/context only; absent independent non-price evidence, keep the arc contested and await a clear on-chain signal (COT/production/inventory/IV/fund flows) before taking a directional position. | +1.4% | 2026-08-17 |
Contested 26
| Narrative arc | ETF | Direction | Confidence | Thesis | Excess vs SPY | First seen |
|---|---|---|---|---|---|---|
| BoJ Yen Normalization boj_yen_normalization | TIP iShares TIPS Bond ETF | Long | 0.86 | Maintain a medium-term long on TIP. The causal chain remains: BOJ/JPY stabilization → lower probability of Japan-driven Treasury sell-offs → less upward pressure on nominal/real yields → relatively favorable for TIPS. No new intervention headlines appeared this refresh, but independent non-price quantitative evidence continues to support that chain: (1) net creation flows have strengthened — +$150M since 2026-08-14 (+0.99% AUM) and +3.41% AUM over 30 days, indicating real buying pressure; (2) Treasury auction demand remains strong (recent bid-to-cover ~2.67), which cushions long-rate spikes; (3) rate structure (10Y=4.69%, 10Y real=2.35%, inflation expectations ≈2.34%, 2s10s=0.5) preserves TIPS' relative case. Price evidence: arc-window excess return is -2.40% (TIP -2.40% vs absolute 0.00%), 8-day window cum +0.61% with max drawdown -0.36% — the market has not fully priced this thesis but also hasn't confirmed it, limiting conviction. Counterweights: the arc's historical analogs remain skewed bearish (42 events: 29 bearish vs 11 bullish) and system liquidity is still contracting over short horizons (Fed net liquidity 4w down $125B). On balance, the independent flow + auction signals justify a modest strengthing of the long stance (confidence moved from 0.82 to 0.86), while retaining caution around history and liquidity risks. | -3.8% | 2026-06-14 |
| US-Iran 2026 War us_iran_2026_war | OIH VanEck Oil Services ETF | Long | 0.84 | I acknowledge the prior view: I previously recommended a mid-term long on OIH (0.82, contested) based on the causal chain that a U.S. blockade/rotation increases the likelihood of sustained shipping-lane disruption → near-term physical tightness and higher insurance/charter premia → futures curve backwardation → higher upstream dayrates/capex → oilfield services benefit (OIH). Today's incremental evidence is the Journal quoting officials (and Hegseth) saying the blockade could be held “indefinitely” with ship rotation — a direct reinforcement of that causal chain. Non-price, asset-specific signals supporting this include: CFTC crude net long = +87,479 contracts (report 2026-08-18) and WTI near-month vs 12-month slope = +17.266% (clear backwardation). Price/context signals: arc window OIH -2.80% vs XLE +11.72% → OIH excess -14.52% (shows relative underperformance); 9-day etf_window cum +0.01% with max drawdown -2.64% (some short-term catch-up but not a regime break). Contra evidence: prediction markets show high short-term ceasefire probabilities (polymarket yes_prob for ceasefire through 8/22 and 8/31 = 1.00 / 0.92), which lowers the probability of a prolonged, economy-wide supply shock; OIH 30d net flows = -8.66% AUM and IV percentile = 4% point to funding/headline complacency risks. Verdict: the new report strengthens the shipping-lane disruption blueprint, so I keep a long view and modestly raise conviction (0.82 → 0.84), but remain contested — position size should be managed and monitored against next CFTC, EIA stock reports, insurance/charter-rate moves and drill-rig/dayrate datapoints. | -8.3% | 2026-05-14 |
| ECB Eurozone Rate Cycle ecb_eurozone_rate_cycle | TLT iShares 20+ Year Treasury Bond ETF | Long | 0.75 | Maintain a medium-term long (1–3 months). The causal chain remains: real ETF net creations → buying/creation demand for long-dated Treasuries → auction/spot absorption of supply → downward pressure on long yields → positive for duration (TLT). Fresh non-price quant evidence independently supports this chain: TLT has continued real net inflows (since 2026-08-14 net +$824M; AUM $46.9B; 30-day inflows +9.28% AUM), indicating persistent cash buying; Treasury auction demand is strong (recent 7Y bid-to-cover mean ≈2.67), implying supply is being absorbed; CFTC 30y net = -361,383 contracts (reporting week 2026-08-18) remains net-short but has retraced slightly (week-on-week +3,441), consistent with partial short covering and squeeze risk. Price facts: arc-window cumulative TLT -4.10%; recent 9-day etf window cum +0.34% with max DD -1.5%; a 4-day price_in excess = +2.52% (recent short-run outperformance). Options and vol are neutral-to-cautious: ATM IV percentile 40%, put-call skew +0.82pt, MOVE=73.4 (calm). Synthesis: independent, non-price flows + auction absorption + COT dynamic reinforce the long thesis; price has not fully realized the thesis so conviction is increased modestly but key risks (high nominal 10Y, still-large net shorts, upcoming US core PCE/GDP) warrant continued monitoring. | -7.7% | 2026-07-18 |
| RBA Australia Rate Cycle rba_australia_macro_policy | TIP iShares TIPS Bond ETF | Long | 0.65 | Maintain a medium-term long on TIP. The prior assessment (2026-08-19, long, conf=0.60) judged the triggering news (UK mining / Q2 growth) not to be a direct RBA hawkish shock and relied on structural evidence: sustained fund flows into TIP, implied rates and low bond volatility pricing a benign/softer rate path, and solid auction demand. I re-ran adversarial checks: (1) the non-price reference metrics still hold or are slightly more supportive — 30-day net creations/flows = +3.77% AUM (as of 2026-08-20), implied Fed funds from futures = 3.63% (still pricing easing/flat vs earlier ~3.72%), MOVE = 73.4 (low), 10Y real rate = 2.35% (down from ~2.44%), and auction bid-to-cover ~2.67 (strong demand); (2) price signals are mixed but not decisive: price_in_excess (3d) = -0.35% while the 9-day ETF window shows cumulative +0.62% and max drawdown -0.36% — per rules price is a tape cross-check only; (3) the event itself (UK mining/UK Q2) lacks a causal chain to RBA policy, so it is not a structural counterexample. Taken together, at least one non-price quantitative line (flows, implied rates, MOVE, real yields, auction demand) supports TIP; therefore I modestly raise conviction to 0.65. A change of view would require direct, non-price evidence of a hawkish Australian policy inflection (e.g., AU CPI surprise persistently above expectations, RBA hawkish guidance, or other arc-specific non-price indicators). | +0.1% | 2026-08-19 |
| RBA Australia Rate Cycle rba_australia_macro_policy | TLT iShares 20+ Year Treasury Bond ETF | Long | 0.60 | Acknowledging the prior (2026-08-19) assessment: it recommended a modest long TLT (confidence 0.50) based on market-implied easing (Fed funds futures ~3.72%), extreme CFTC net-short positioning in long-dated Treasuries, and persistent ETF inflows. Re-checking those reference classes now: the non-price evidence still supports the long case — CFTC treasury_30y net short remains extreme at -361,383 contracts (four-period trend moving from -389,522 → -361,383; percentile=100), market-implied Fed funds path has eased further to ~3.63% (still pricing cuts), TLT AUM rose to $46.9B with net creations +$824M since 2026-08-14 (+1.76% AUM) and 30d inflows +9.28% AUM; treasury auction bid-to-cover avg ~2.67 indicates strong demand. Price cross-checks show recent 4d price_in excess +2.52% (TLT +1.22% vs benchmark -1.30%), 9d window cum +0.34% and max dd -1.5%; arc cumulative remains TLT -2.82%. These non-price signals continue to point to a structural tailwind for long-duration positions (crowded short + flows + easing priced), so I keep the long bias and modestly strengthen conviction to 0.60. Key risks remain: 10y nominal/real yields are still elevated (10Y≈4.69% / real≈2.35% / term premium≈0.84%); sustained hawkish macro surprises, weakening auction demand, or liquidity tightening would invalidate this stance. The immediate trigger (UK Q2/mining news) is largely orthogonal to the US Treasury drivers and is neutral for this thesis. | -9.8% | 2026-08-19 |
| Taiwan-China Strait Tension taiwan_china_strait_tension | TLT iShares 20+ Year Treasury Bond ETF | Contested | 0.55 | Prior (2026-08-21) I recommended a medium-term long in TLT (long, confidence 0.65, contested) based on non-price evidence: real-creation inflows, partial short covering and solid auction demand. In this daily refresh the conclusion remains contested: there are continued non-price supports but also clear countervailing signals. Supports: TLT saw real net creations +$512M since 2026-08-17 (+1.11% AUM), 30-day cumulative inflows +9.29% AUM (continued funding demand); CFTC treasury_30y net = -361,383 contracts with weekly change +3,441 (small short covering); treasury auction bid-to-cover recent mean ≈2.67 (strong demand). These sustain the chain (institutional inflows → duration demand → ETF support). Counters: 10Y nominal = 4.69%, 10Y real = 2.35%, term premium = 0.8393% (long-end yield pressures from supply/uncertainty); Fed net liquidity 4-week change -$125B (mild liquidity tightening); prediction markets price a low probability for China–Taiwan military clash (yes_prob ≈0.07–0.08), weakening the geopolitical-safe-haven transmission. Price evidence: arc price_in excess = -5.42% (TLT underperformed benchmark by -5.42% in arc window), etf_price_window (8d) cumulative = -1.42%, max drawdown = -1.70% — prior downside has not been fully reversed. Because non-price evidence points both ways, I keep a contested stance. Execution: remain cautious, use small position size and tight stop, monitor申赎 flows, COT and long-end yield/term-premium/auction signals for the next decisive move. | -8.0% | 2026-06-18 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | TIP iShares TIPS Bond ETF | Long | 0.55 | Acknowledge prior assessment (2026-08-23): it kept a medium-term long on TIP (confidence=0.60, contested), arguing that if macro/market expectations pivoted dovish or terminal rates were revised down, real rates and long nominal yields could fall and TIPS would benefit; prior also cited TIP net inflows as supportive. Critique of prior: it gave too much directional weight to flows/financing and did not sufficiently decompose financing channels (financing by itself is directionally ambiguous per rules). Update vs prior: the trigger is an ai_compute_capex_race event (a semiconductor startup raising ~ $2bn). Per policy, such financings do not provide a robust causal chain to higher CPI or a Fed dovish pivot—capex can raise demand but also signals supply/expansion and dilution. Quantitative evidence and impact: price-derived: TIP arc excess return −2.41% (window cumulative −2.41%, max drawdown −2.85%, latest close 107.13) shows recent pressure; non-price signals are mixed—supportive: TIP 30d net inflows +3.84% AUM (allocation/flow support); constraining: 10y real yield ≈ 2.35% (directly negative for TIPS valuation) and market-implied Fed funds path ZQ=3.735% (not pricing a clear easing). MOVE=73.4 (low volatility) and strong treasury auction demand (bid-to-cover ~2.67) are neutral-to-mildly supportive. Conclusion: structural thesis for TIP remains long (TIPS benefit if real yields fall), but the semiconductor financing event is neutral/ambiguous for inflation/policy; given persistent high real yields and only mixed non-price evidence, maintain long but lower confidence further (0.60 → 0.55). Change_vs_prior = weaken. | -3.9% | 2026-06-03 |
| EU-US Tariff Dispute eu_us_tariff_dispute | CHPS Defiance Semiconductor Equipment ETF | Long | 0.55 | Maintain long. The prior view reasoned that the immediate tariff-risk pathway (tariff_policy_shock → sector repricing) had been weakened and therefore recommended a controlled long in CHPS. Prior non-price support comprised (1) DRAM spot supply tightness (TrendForce sample ~+5%) and (2) a material decline in CHPS short interest (−18.1%, days-to-cover=1). Since then, price metrics show further underperformance: current price_in excess = −5.64% (36d window: CHPS −14.96% vs SMH −9.32%, flagged=True); etf_window cumulative −14.96%, max drawdown −25.7%, latest close 82.6384 (2026-08-20). Crucially, the non-price industry signals that underpinned the bullish case have not been invalidated — DRAM spot trend has strengthened to +9.16% (TrendForce), reinforcing storage-chain fundamentals, and CHPS short interest (settlement 2026-07-31) remains indicative of reduced short pressure (4,138 shares, −18.1%). However, the arc’s historical analogs remain weak (many event_opp are first_signal/unclear), and CHPS’s small AUM and low volume (today $0.8M vs 20d $3.0M) mean execution/liquidity risk. Therefore I keep the prior controlled long stance: the non-price supply/positioning evidence supports long, but price underperformance and liquidity constrain position sizing and warrant phased entry. | -17.3% | 2026-08-19 |
| ECB Eurozone Rate Cycle ecb_eurozone_rate_cycle | UUP Invesco DB US Dollar Index Bullish Fund | Short | 0.46 | Maintain a mid-term short on UUP (betting on dollar weakness). The prior transmission logic still holds: (1) euro-facing hard data remain strong — Germany manufacturing PMI = 54.1 (σ=2.0) and Eurozone manufacturing PMI = 52.8 (σ=1.2), providing non-price fundamental support for EUR appreciation; (2) fund flows continue to show dollar unloading — recent net redemptions have widened (since 2026-08-10 roughly $100M out, ≈-31.03% AUM; 30d cumulative -32.38% AUM), an independent non-price signal of dollar sell pressure. Price has already reflected part of the move (arc window price_in excess = +2.01%; window cum +2.01%; max peak +4.57%; max drawdown -2.52%), but price is corroborative not the primary driver. Main counterarguments are short-position crowding (FINRA short interest +22%, days-to-cover=3.22), and mixed historical responses to hawkish-rate-type events (e.g., event_opp #205 saw UUP 4d excess = -2.50%). On balance the newly available non-price signals (larger ETF outflows and persistent strong EU PMIs) reinforce the prior euro-strong / dollar-weak thesis, so I strengthen the short stance modestly while keeping tighter position sizing and stop discipline due to crowding and partial price realization. | -1.6% | 2026-07-18 |
| US CRE / Office Crisis Cycle us_cre_office_crisis_cycle | XLRE Real Estate Select Sector SPDR Fund | Contested | 0.45 | Prior (2026-08-20) argued short (conf=0.55) because a structural underwriting/insurance channel (climate risk being priced into underwriting) would raise insurance/holding/retrofit costs and/or cap rates, reducing office CRE valuations. That view relied on non-price signals at the time (net creations/redemptions outflows ~-$150M since 8/11, rising FINRA shorts, and tighter 4‑week Fed liquidity). Since then there are no new event_opps—the structural channel remains plausible—but the fresh independent datapoints are mixed. Supportive: continued net ETF outflows since 2026-08-12 of -$155M (-1.82% AUM) and persistent FINRA short interest (7,454,300 sh, +16.4% over 7 periods, days‑to‑cover=1.56) keep the repricing/positioning worry alive. Contradictory: price and volatility do not confirm further downside—price_in_excess (6d)=+0.95% (XLRE -0.42% vs SPY -1.37%); etf_window 6d cumulative -0.42% with max DD -1.41%; option IV is at low historical percentile (call IV 14.36% / put IV 10.3%; 5d IV change -0.56pt; skew more negative and cheaper relative to SPY), implying low market-paid downside insurance. System liquidity also eased vs last week (Fed net liquidity 4w moved from -$191B → -$125B, WALCL−TGA−RRP=$5,792B). Because these non-price signals conflict (flows/shorts vs IV/price/liquidity) and no new structural event has strengthened the causal chain, the assessment becomes contested rather than a clear short; confidence is reduced to 0.45. Historical analog strength remains weak (first_signal); the arc is still early and not past inertia half‑life, so do not treat prior signal as decayed yet. | +0.9% | 2026-08-14 |
| US-Iran 2026 War us_iran_2026_war | TIP iShares TIPS Bond ETF | Long | 0.40 | Maintain a mid-term (1–3 month) long on TIP. The direction rests on non-price quantitative chains rather than price alone: (1) demand-side flows show institutional buying—30-day cumulative net inflows +3.84% AUM and net creations since 2026-08-17 of +$161M (~+1.07% AUM), indicating continued demand support; (2) auction-side demand is strong (recent bid-to-cover mean ≈2.67), reducing near-term upside risk to nominal yields and supporting TIPS relative attractiveness; (3) the market-implied Fed funds path is neutral-to-slightly-dovish (implied Fed funds ≈3.735%) and MOVE is subdued at 73.4, which together limit the odds of a sharp rise in real yields. Price signals show arc price_in excess = -2.07% (TIP underperformance), but the 9-day etf_price_window cumulative +0.22% with max drawdown -0.36% indicates a small recent recovery and no structural breakdown of the non-price evidence. Offsetting risks: prediction markets price a high short-term ceasefire probability (polymarket yes_prob 0.92/1.00), removing some geopolitical safe-haven premium; and the arc excess negative suggests partial re-pricing has already occurred. Net: the non-price evidence still favors long, so keep the call with moderate-low confidence (0.40) and maintain contested status. | -4.6% | 2026-07-21 |
| AI Compute Capex Race ai_compute_capex_race | DTCR Global X Data Center & Digital Infrastructure ETF | Long | 0.38 | A frank critique of the prior view: the 2026-08-21 assessment maintained a long stance despite realized negative price performance in the arc window, arguing that independent non-price signals (large AUM inflows and DRAM spot strength) supported the AI compute capex thesis and raised confidence to 0.38. Today's refresh: keep the long call. The transmission chain remains: real money allocation toward the theme (AUM flows) + evidence of DRAM tightness → greater likelihood of continued capex/ procurement by hyperscalers and chip buyers → beneficiaries inside DTCR. Price context is important but not dispositive: arc window price performance is negative (DTCR cumulative -12.02%, DTCR vs SPY excess -13.54%, max drawdown -18.18%), indicating much downside has already occurred and price should be treated as reflection rather than primary directional evidence. Independent non-price signals still support the bullish thesis — AUM $2.2B (as of 2026-08-20) with 30d net flows +9.13% AUM and 90d +8.76% AUM shows continuing allocation into the ETF; TrendForce DRAM spot indices show material spot strength (main contract +9.16%, DDR5_16Gb +13.85%), consistent with supply tightness that would spur capex/purchasing. Offsetting risks persist: FINRA short interest 548,188 sh (+171%, days-to-cover=1) implies crowding/short-accumulation risk, and Fed net liquidity has contracted (-$125B 4-week). No new event_opp has appeared to flip the thesis. Therefore maintain long with moderate confidence and event-driven sizing; watch near-term macro releases (US core PCE/GDP 2026-08-26) and upcoming company earnings as potential triggers. Note: price-derived metrics are used for cross-checks only; direction is grounded on non-price quantitative signals listed above. | -13.5% | 2026-06-03 |
| Clean Energy Transition clean_energy_transition | XLB Materials Select Sector SPDR Fund | Contested | 0.35 | The prior short thesis (anchored on a US retail sales downside surprise plus crowded speculative longs in copper) has not gained new confirming non-price evidence over the past four days. Price and flow behaviour are mixed: XLB rose 2.49% over the arc window while SPY fell 0.90%, producing a XLB−SPY excess of +3.39% (price_in excess); the 9-day etf_window cumulative return is +1.04% with max drawdown −1.75% (signalling partial price-side realization). Non-price signals conflict: FINRA short interest increased to 15,040,364 shares (+8.8%, days-to-cover=1.37), indicating active short positioning (and squeeze risk); 30‑day net flows are −1.62% AUM (net outflow), which mildly supports the bearish case; Fed net liquidity declined ~ $125B over 4 weeks (small tightening). Offsetting these, European manufacturing PMIs—notably Germany 54.1 (σ=2.0)—are materially stronger, supplying independent demand-side evidence that weakens the original US-retail-driven bearish chain. Options IV is at a 0% recent percentile (cheap protection); call IV 20.17% vs put IV 17.04% (skew −3.12pt) is useful for sentiment context but not a standalone directional input. Historical analog strength is weak (first_signal). Given the conflicting non-price evidence and partial price realization, I keep the arc as contested and reduce confidence from 0.45 to 0.35 (no strengthen — no new independent non-price confirmation). | +3.4% | 2026-08-17 |
| AI Compute Capex Race ai_compute_capex_race | XLK Technology Select Sector SPDR Fund | Long | 0.30 | Maintain a cautious medium-term (1–3 month) long tilt in XLK, with a small increase in conviction. The prior view (2026-08-18, long, confidence=0.20) rested on: AI training/inference driving sustained demand for compute and memory → hyperscalers and semiconductor/cloud constituents benefit. Critique of the prior view: although DRAM spot had already been cited as non-price support, prices and flows had not validated the thesis (price_in excess = -5.86%; arc cumulative XLK -4.42%; window max drawdown -13.31%), so the downshift to 0.20 was reasonable. Since that assessment, a direct non-price demand signal strengthened the causal chain — DRAM spot trends rose further (DRAM index +9.16%; DDR5_16Gb now $37.0, up ~13.85% in the period and materially above the prior cited $34.2), which maps clearly to AI-driven memory demand (AI training → more HBM/DDR5 demand → higher DRAM spot → positive for storage/semiconductor constituents and thus XLK). Accordingly I raise conviction modestly to 0.30 (constrained by the daily-refresh +0.10 cap). Offsetting signals remain: XLK net redemption since 2026-08-13 of -$188M (-0.16% AUM) and 30d net flow -0.30% AUM; option IV sits at the 0th percentile vs recent history (cheap puts/protection); put-call skew -1.55pt (more negative than SPY’s -1.16pt); and price action still shows underperformance (price_in excess -5.86%). In sum, the stronger DRAM spot move is an independent, arc-relevant non-price reinforcement that justifies a small confidence increase, but material liquidity/positioning and price underperformance keep the view cautious and active monitoring of earnings (NVDA 8/26, AVGO 9/02) and flow/positioning data is required. | -5.9% | 2026-06-13 |
| US-China Tech Decoupling us_china_tech_decoupling | SMH VanEck Semiconductor ETF | Short | 0.28 | Maintain a medium-term bearish tilt on SMH but with reduced conviction. The primary non-price directional channel remains intact: the FOMC minutes note that inflation is still elevated, supporting a tightening/hawkish policy pathway; combined with a measured decline in system liquidity (WALCL−TGA−RRP = -$125B over the past 4 weeks), this creates the classic rate→valuation pressure channel for growth-sensitive semiconductors (consistent with rate_decision_hawkish_surprise). However, several non-price structural signals weaken the timing and aggressiveness of a short: (1) industry fundamentals — DRAM spot shows material strength (~+9.16% on TrendForce), implying supply-demand support for parts of the chain; (2) options/volatility measures show low priced protection (SMH ATM IV at 4th percentile), reducing immediate hedging premia; (3) flows/positioning are mixed — net ETF creations/redemptions since 2026-08-14 show net outflows of ~$1.846B (-2.76% AUM), while short-interest reported fell ~31.4% to 11,105,545 shares (shorts partly covered, lowering forced-cover risk). Price evidence is mixed and merely supportive context: arc-window (2026-06-16→08-18) SMH cumulative -7.50% vs SOXX -10.12% (SMH excess +2.62%); 46-day price_in excess = +3.02% (SMH -9.02% vs SOXX -12.04%); short-term 6d excess = -3.29% (SMH -4.66% vs SOXX -1.37%). Net: the macro/liquidity channel still justifies a short bias, but industry strength, low IV and changed positioning reduce the edge — keep direction = short but downgrade conviction and avoid adding size until clearer non-price deterioration or a tactical re-test of the thesis. | -11.1% | 2026-08-15 |
| US-Iran 2026 War us_iran_2026_war | AMLP Alerian MLP ETF | Long | 0.28 | Maintain a medium-term lean-long on AMLP with tight position sizing and strict risk controls. Non-price evidence still partially supports the transmission: commodity tightness → higher midstream throughput/fee capture → stronger MLP cashflows. Key 'on‑point' signals: CFTC crude net-long +87,479 contracts (reporting 2026-08-18, elevated over recent weeks) and AMLP net creations of +2.60% AUM since 2026-08-10 (fund-level demand). Historical analogs in the arc favor bullish outcomes (249 bull vs 52 bear ≈79%), adding consistency. Offsetting risks: AMLP has materially underperformed XLE over the window (price_in excess = -7.94%; AMLP window cum = +3.71% with max drawdown -8.11%), option skew shows relatively more put demand (AMLP put−call +2.61pt vs XLE −2.49pt, diff +5.10pt), ATM IV is at low historical percentile, FINRA short interest +16.2% (days-to-cover 4.06) and HY OAS 2.75% indicate crowding/credit constraints. Net: direction remains long because of on‑point non-price support, but conviction is low and position sizing should stay conservative given competing de‑risk signals. | -0.6% | 2026-07-18 |
| ECB Eurozone Rate Cycle ecb_eurozone_rate_cycle | TIP iShares TIPS Bond ETF | Long | 0.25 | The prior assessment (2026-08-23) set a long stance with contested status (confidence=0.25), arguing that price and structural signals were mixed: the arc window still shows an excess loss (TIP excess = -2.25%) while independent non-price indicators (net inflows, modest easing in real yields, slight breakeven increase) supported TIP. I agree that the prior reliance on flows and real-yield signals was reasonable and caution that price alone should not overturn the thesis—per rules, direction must be grounded in non-price quantitative evidence. In this daily refresh, with no new event triggers, the independent non-price evidence remains supportive: TIP AUM continues to net inflow +$161M (since 2026-08-17, +1.07% AUM); 30-day cumulative inflows +3.84% AUM; FRED shows 10y real yield = 2.35% and 10y breakeven = 2.34% (a modest easing of real-yield pressure on TIPS). Price facts: arc-window excess = -2.25%, but the recent 9-day etf_price_window cumulative +0.22% with max drawdown -0.36% and last close 107.13 (2026-08-21) show short-term recovery that does not materially contradict the prior case. Therefore, absent a new independent non-price signal pointing the other way (e.g., sudden COT rotation, weak auction demand, or renewed rise in real yields), I keep the prior long position and confidence. Note: near-term US core PCE/GDP releases are material event risks that could force rapid re-evaluation. | -6.0% | 2026-06-11 |
| ECB Eurozone Rate Cycle ecb_eurozone_rate_cycle | AGG iShares Core U.S. Aggregate Bond ETF | Long | 0.22 | Acknowledging the prior (2026-08-19) view: I previously positioned AGG as a tactical long (confidence=0.30, contested), primarily supported by non-price signals: market-implied easing priced into Fed funds (ZQ futures), robust US Treasury auction demand, and ETF net inflows—these form the structural case for duration exposure. Current conclusion: maintain the long stance but further reduce conviction. The non-price supports remain: implied Fed-funds path at 3.63% (ZQ), recent Treasury bid-to-cover mean ≈2.67 indicating continued demand, and AGG net inflows +$68M since 2026-08-14 (+0.05% AUM) with 30d cumulative +0.38% AUM—consistent with the chain (policy pivot/pricing → Treasury demand/price support → ETF flows). Offsetting factors that prompt a confidence downgrade: the thesis is partly price-realized (price_in excess = -5.34%; AGG window cumulative -1.55%; max drawdown -2.14%), and option markets are pricing elevated downside protection (ATM IV: call 4.64% / put 8.13%; IV percentile 92%; put-call skew +3.49pt), while the long end remains high (10Y=4.69%; 2s10s=0.50; term premium=0.8393%). No new ECB-specific directional event has appeared in this refresh, so I keep the direction long but weaken conviction from 0.30 to 0.22. | -5.3% | 2026-06-11 |
| BoE UK Gilt Cycle boe_uk_gilt_cycle | PFF iShares Preferred and Income Securities ETF | Contested | 0.20 | Summary: The prior view was a weak short (confidence 0.15) based on elevated term premium/long yields, unrepaired credit spreads and PFF flows/short accumulation. Those non-price quantitative supports remain: 10Y=4.69%, 2s10s=0.50, HY OAS ≈2.75%, PFF 30d net flows ≈ -1.78% AUM, FINRA short shares +11.3% (days-to-cover=1.85). Price-derived evidence also shows relative weakness: price_in excess (15d) = -1.59%, etf_window cum = -0.52%, max drawdown = -1.42%. These structurally favor a short stance. The current triggering event (event #3759 — UK growth tipped to slow; mapped to central_bank_dovish_surprise / central_bank_intervention_fx) presents a plausible transmission: weaker growth → more dovish BoE forward pricing → lower gilt yields → relative relief for income/credit-sensitive ETF such as PFF. However, the event is currently a narrative-level dovish signal (timeline flagged as first_signal, i.e., weak historical-analog strength) and lacks corroborating, arc-specific non-price quantification (no provided UK swap-implied rate re-pricing, no confirmed gilt yield decline). Therefore, while structural non-price evidence remains biased to the downside, the new event introduces a credible counterargument that is not yet validated quantitatively. I therefore label the arc as contested — awaiting direct, arc-specific non-price confirmations (UK rate forwards/gilt supply demand, BoE guidance, or credit spread compression) before committing to a directional flip. | -1.6% | 2026-08-01 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | AGG iShares Core U.S. Aggregate Bond ETF | Long | 0.17 | Maintain a small-duration long (long) with low confidence and contested status. Prior assessment (2026-08-23) set the position long, confidence=0.18, contested, on the premise that there was no coherent, non-price dovish evidence (e.g., an implied fed-funds path move down ≳20bp or clear Fed signaling) to justify stronger duration exposure—while price action had already reflected duration stress (arc window AGG cumulative -0.84%, AGG vs SPY excess -4.42%, AGG max drawdown ≈ -2.14%). Critique of the prior view: the prior standard remains valid — absent a policy-path or macro-data pivot, price moves alone should not flip the arc direction. Since the prior note there is no new on-path dovish evidence: ZQ-implied fed funds ~3.735% (no material downward repricing), prediction market yes_prob ≈0.47 (no clear cooling), 10Y = 4.69% and 2s10s = 0.50 — these do not signal a dovish pivot. Offsetting evidence supporting bonds includes AGG 30d net flows +0.42% AUM, recent treasury auction bid-to-cover ~2.67 showing demand, and MOVE = 73.4 (no wholesale selloff). Options show increased demand for downside protection (put IV 8.13% vs call IV 4.64%; IV percentile 92%). Price short-window is neutral: AGG 9d cum +0.03% / max drawdown -0.48%; arc window still AGG -0.84% (AGG-SPY excess -4.42%). Conclusion: no on-arc non-price evidence has arrived to overturn the cautious small-long stance, so I keep long at low conviction (0.17), awaiting US core PCE/GDP or explicit Fed signaling to materially change the view. | -4.4% | 2026-06-07 |
| US-China Tech Decoupling us_china_tech_decoupling | FXI iShares China Large-Cap ETF | Short | 0.16 | Maintain a medium-term short on FXI, but with materially lower conviction. The structural channel remains “Beijing regulatory/tighter scrutiny of cross‑border tech investment → re‑pricing of China large‑cap tech” (dominant blueprints: regulatory_action_chinese_tech, policy_inflection_export_control), and the arc continues to register multiple regulatory event_opps (majority bearish). Non‑price quantitative support for a short position: FINRA short interest at 71,582,778 shares (+0.9% vs prior; +72.4% over 7 periods; days‑to‑cover = 3.25) and 30‑day net redemptions of −1.09% AUM indicate market participants are positioned for/withdrawing from the ETF. Offsetting risks: the move is significantly price‑in (FXI vs EEM excess +13.12%; 43‑day ETF window cumulative +7.27%, max drawdown −5.5%; latest close 35.86), options IV is low (IV percentile 29%), and EM volatility (VXEEM = 27.05, z = −2.06) is unusually calm — these factors increase squeeze/realization risk. Fed liquidity has tightened modestly (WALCL−TGA−RRP = $5,792B; 4‑week −$125B) and FOMC minutes highlight inflation concerns that can feed risk premia, but that is a macro channel, not a direct China‑regulatory read. Net: structural/regulatory evidence still leans negative for FXI, but crowding/price_in force a lower allocation and smaller conviction (confidence lowered to 0.16). | +4.4% | 2026-06-20 |
| US-Iran 2026 War us_iran_2026_war | SLV iShares Silver Trust | Contested | 0.16 | Remain contested. The reported U.S. discussion of an indefinite blockade logically strengthens the conflict → shipping/insurance → safe‑haven demand blueprint that would support precious metals (silver’s safe‑haven component). Non‑price quantitative evidence is mixed: direct supports include CFTC net long +11,695 contracts (90th percentile), 10Y real rate = 2.35% (20d -8bp, easing holding cost), SLV 30‑day creations +1.02% AUM (AUM $34.1B) and option IV at the 80th percentile—these are asset‑specific, non‑price drivers consistent with increased positioning and risk premia. Offsetting this, prediction markets still ascribe high probability to near‑term ceasefire/de‑escalation (polymarket yes_prob for ceasefire through Aug31 = 0.92), and price has already discounted some risk (arc excess return -13.25%; 55d cum -5.27%; max DD -24.77%). COT at extreme levels also signals crowding and drawdown susceptibility. The new event reinforces the escalation → safe‑haven causal path but does not resolve the strong, non‑price evidence pointing toward short‑term de‑escalation and crowded longs; therefore no directional flip or confident tilt. | -6.8% | 2026-06-03 |
| NATO / EU Defense Spending Cycle nato_eu_defense_spending | PPA Invesco Aerospace & Defense ETF | Long | 0.12 | Maintain a moderate-term long bias on PPA, but reduce conviction further. The prior assessment leaned long citing the defense-spending causal chain (Orion26 → higher EU/NATO procurement) plus recent ETF inflows; I previously criticized that view for relying too heavily on flows and narrative rather than independent non-price triggers. Since the prior note there has been no new structural event or concrete EU/NATO budget commitment to justify adding risk. Price has continued to price in part of the thesis: price_in excess = -4.29% (5d PPA -5.19% vs SPY -0.90%), etf_window 5d cumulative -5.19% and max drawdown -5.19% (close 173.67 on 2026-08-21), indicating partial realization. Non-price signals are mixed: AUM still shows net inflows (since 2026-08-10 +$33M = +0.40% AUM; 30d +0.76% AUM), a mild demand confirmation; Eurozone/Germany manufacturing PMI surprised to the upside (EU manuf 52.8 vs 51.8, σ=1.2; DE manuf 54.1, σ=2.0), which supports industrial capacity and fiscal room as background but is a weak, lagged conduit to immediate defense procurement. Prediction markets still show low near-term NATO×Russia clash odds (Polymarket 8/31 yes_prob=0.04; 12/31 yes_prob=0.24). Short-interest dynamics (short shares 119,035; 7-period cumulative +152.7%; days-to-cover=1) indicate crowding and squeeze risk but are not bullish demand evidence. Conclusion: structural thesis not falsified but lacks fresh independent non-price confirmation while price has further realized gains; lower conviction from 0.20 to 0.12 and keep a small long stance pending explicit non-price catalysts (e.g., formal NATO/EU procurement/budget announcements or sustained rises in prediction-market odds). | -4.3% | 2026-08-15 |
| ECB Eurozone Rate Cycle ecb_eurozone_rate_cycle | SOXX iShares Semiconductor ETF | Long | 0.12 | Maintain a medium-term (1–3 month) lean-long on SOXX. The structural bull case—ai_compute_capex_race driving sustained capex into high-end semiconductors for cloud/AI—remains intact. Systemic liquidity is still large (WALCL−TGA−RRP = $5,916B, 4-week +$103B) and recent Eurozone PMIs surprised to the upside, providing a supportive macro backdrop. However, several independent non-price signals and price action have reduced conviction: (1) material fund redemptions for SOXX since 2026-07-22 of -$1,698M (-3.87% AUM); (2) options market skewing toward puts (put IV 64.48% vs call IV 58.66%; skew +5.82pt; delta vs QQQ skew = +8.13pt), indicating higher demand for downside protection; (3) realized price deterioration — SOXX -11.23% over the arc (2026-07-13→07-28) with SOXX-QQQ excess -6.13%, and a 9-day etf_price_window cum -5.37% / max DD -13.9%. The structural bull thesis survives but is weakened by flows and risk-pricing, so I remain long at a lower confidence (0.12 vs prior 0.18). | -8.3% | 2026-07-12 |
| Trump 2026 Tariff Policy trump_tariff_2026_policy | SMH VanEck Semiconductor ETF | Contested | 0.02 | Maintain a contested stance on the Trump 2026 Tariff Policy arc for SMH. Critique of prior view: the previous assessment correctly noted there was no quantitative policy evidence that would close the causal chain (corporate lobbying → policy execution → directional impact on SMH); that remains true today (no USTR/Congress/prediction-market move). Fresh non-price signals are mixed and offsetting. On the positive side, semiconductor fundamental data show clear supply tightness: DRAM spot (main) +9.16% and DDR5_16Gb +13.85% (TrendForce, as of 2026-08-21), which would logically transmit as tighter memory market → support for memory suppliers and related SMH holdings. On the negative side, ETF-level liquidity signals show significant redemptions: net creations/redemptions since 2026-08-14 = -$1,846M (-2.76% AUM), 30d net -0.43% AUM, indicating investor withdrawals and increased crowding/withdrawal risk. Price-derived cross-checks show only a small SMH excess vs SOXX (+0.19% excess; SMH -1.58% vs SOXX -1.76% from 2026-08-10 to 2026-08-21), 10-day cumulative -1.58% and max drawdown -5.66% (close 560.42 on 2026-08-21), so the market has not clearly priced a directional outcome. Options/volatility show low IV percentile (4%) and a relatively negative skew vs peer (SMH skew ~ -4.28pt vs SOXX), consistent with complacency but not a standalone direction signal. Given the offsetting independent non-price signals and rising narrative heat (higher article count), I keep the arc contested and reduce conviction slightly to reflect redemption risk and crowding. | -0.6% | 2026-08-08 |
| Taiwan-China Strait Tension taiwan_china_strait_tension | CHPS Defiance Semiconductor Equipment ETF | Contested | 0.01 | Maintain a very low-confidence, contested stance (neither adding nor closing position). The prior view (2026-07-22) reduced exposure to a tiny long (confidence 0.005) because, although two structural blueprints (taiwan_strait_geopolitical_tension and ai_compute_capex_race) remained relevant, there was little independent non-price evidence and prices had already priced in much of the prior optimism (arc window CHPS cumulative -22.69% with CHPS-SMH excess -6.48%), weakening conviction. Today’s refresh shows: (1) price/window data still bear that out (etf_window cumulative -22.69%, max drawdown -32.74%, latest close 82.987; price_in excess = -6.48%), which supports the prior claim that much has been priced in; (2) new independent non-price signals are mixed — DRAM spot prices are materially higher (DRAM basket +9.16%, DDR5_16Gb +13.85%), lending causal support to the ai_compute_capex_race thesis (tight memory supply/demand → benefit to memory-related holdings), while prediction markets still indicate a low probability of a China-Taiwan invasion (yes_prob ~0.03–0.09, typical ~0.07–0.08), which weakens the near-term geopolitical trigger. Volatility is low (VXN=23.26, z=-1.04) and Fed net liquidity has edged down ($5,792B, 4-week -125B). Because independent signals pull in different directions and price has largely reflected earlier moves, there is no clear basis to upgrade or flip the prior call — remain contested with confidence unchanged at 0.005. | -25.6% | 2026-06-22 |
Decaying (signal fading, top 50, click to expand)
| Narrative arc | ETF | Direction | Confidence | Thesis | Excess vs SPY | First seen |
|---|---|---|---|---|---|---|
| US CRE / Office Crisis Cycle us_cre_office_crisis_cycle | VNQ Vanguard Real Estate ETF | Short | 0.70 | Maintain a mid-term (1–3 month) short on VNQ; this refresh strengthens the prior short thesis. The prior argument centered on office CRE repricing: rising underwriting/insurance costs, higher compliance and climate-related spending, plus funding/discount-rate pressure, should raise risk premia and cap rates and compress REIT valuations. Key non-price evidence cited previously remains intact: US retail sales surprised weak (us_retail_sales_mom = -0.6% vs 0.1, σ = -2.8) and Fed net liquidity (WALCL−TGA−RRP 4w) has contracted (4w Δ = -$191B). The new trigger (2026-08-14, “CRE’s ESG Retreat Masks Growing Spending On Climate Risk”) directly reinforces the causal chain: extreme weather and underwriting/regulatory shifts → insurers raise premiums and underwriters embed climate risk → higher operating/insurance costs and higher required yields → cap rates and discount rates move higher → VNQ downside. Price-derived evidence is consistent with this: 28-day price_in excess = -1.86% (VNQ 0.8% vs SPY 2.65%); arc-window cumulative VNQ +1.02% vs SPY +3.63% (excess -2.60%); 28d cum = +0.8% with max drawdown -4.53%, indicating relative weakness with remaining downside to come. Offsetting signals: low VNQ IV percentile (13%) and a -9% drop in reported short interest (days-to-cover = 1.9) imply cheap protection and potential short-squeeze/positioning risk that temper the pace of decline. On balance, the event provides a direct, arc-relevant channel that strengthens the original thesis; confidence is nudged from 0.65 to 0.70. | -1.5% | 2026-08-14 |
| Africa Military Conflict africa_military_conflict | XLE Energy Select Sector SPDR Fund | Long | 0.65 | Maintain a medium-term bullish stance on XLE (1–3 months). The structural, non-price evidence that directly maps to oil is supportive: CFTC crude net speculative long at +79,916 contracts (report week 2026-08-11; net long despite a modest pullback, ~55th percentile) and a pronounced WTI backwardation (near month $86.38 vs 12-month $73.71; slope +17.189%). These two are the correct “asset-specific” signals for the conflict → oil tightness → energy equity transmission channel. Price-based confirmation is strong but remains tape-validation only: 6d price_in excess = +4.62% (XLE +4.18% vs SPY -0.44%, flagged=True); etf 6d window cum +4.18% (max dd -0.16%); arc window (2026-07-15→2026-08-14) XLE +9.58% vs SPY +2.85% (excess +6.72%). Offsetting risks: the triggering event (sanctions/policy on gold in Sudan) is not a direct oil-supply datapoint and is therefore a second-order political escalation channel; XLE has seen recent net redemptions (since 2026-08-13 net -$168M, -0.40% AUM; 30d -1.41% AUM); options IV is relatively low and technicals are overbought. Net effect: stronger evidence from the oil curve and positioning justifies a modest increase in conviction versus the prior assessment, while preserving caution due to non-oil specificity of the event and fund flow weakness. | +11.2% | 2026-08-12 |
| EU-US Tariff Dispute eu_us_tariff_dispute | MCHI iShares MSCI China ETF | Long | 0.65 | Maintain a medium-term long on MCHI (1–3 months). Structural thesis that the EU–US tariff threat is currently more likely to be cooled/not escalated remains intact and unrefuted by today's facts. Price-derived evidence corroborates investor preference for MCHI: price_in excess = 7.52% (36-day: MCHI 7.73% vs EEM 0.21%), etf_price_window cumulative = 7.73% with max drawdown -4.41%, arc window cumulative MCHI +5.62% / EEM +0.30% (excess +5.32%). Per the rules, these are corroborative only. Crucially, non-price quantitative signals support a constructive risk environment: VXEEM = 26.54 (z = -2.27), indicating relatively low EM volatility, and Fed net liquidity = $5,792B (4-week change -$125B) — liquidity remains ample though slightly pulled back. Flow data is neutral (30d AUM flow +0.01%), but short interest is a crowding/risk flag (15,353,069 shares; days-to-cover 5.94; 7-period +19.1% accumulation). The triggering item (event_opp #3502) is a Brazil sanctions/election story and is not causally linked to an EU–US tariff escalation, so it is neutral to the arc; absent a tariff-specific non-price signal that points the other way, the balance of structured evidence modestly strengthens the prior long view. Continue long but monitor macro calendar and any direct EU–US policy moves. | +5.3% | 2026-08-19 |
| EU-US Tariff Dispute eu_us_tariff_dispute | EWG iShares MSCI Germany ETF | Long | 0.65 | Maintain a medium-term long bias on EWG (1–3 months). The prior view (2026-08-19, long 0.60) argued that the EU–US tariff/trade-dispute arc had not meaningfully deteriorated and that price and flows partially reflected but did not negate a long stance (citing price_in excess ≈ +1.87%, arc-window EWG excess ≈ +1.70%, 30d net flows ~ +6% AUM and positive ZEW). Since then, the structured data modestly reinforce that view: price_in excess (36d) has risen to +2.37% (EWG 6.58% vs EFA 4.20%); etf_price_window cumulative +6.58% with max drawdown -4.85% and latest close 43.92; 30d cumulative flows remain net inflow +5.29% AUM (ongoing demand). These price + non-price signals support the long case. Countervailing risks remain — historical analogs within this arc are somewhat skewed negative (sample: 10 bullish vs 14 bearish events), euro speculative net-short persists (COT euro_fx net ≈ -65,198), and option-side risk premia are elevated (ATM IV 17.77%, IV percentile 67%), signaling downside protection demand. The current trigger (#3502, Brazil election/sanctions item) is not causally linked to the EU–US tariff transmission chain and is therefore neutral noise for this arc. Net: keep the long orientation but modestly strengthen confidence from 0.60 → 0.65 because both relative price performance (price_in excess +2.37%) and persistent net flows (+5.29% AUM) corroborate demand; continue to monitor euro positioning and option-implied risk for tail downside and manage exposure accordingly. | +4.6% | 2026-08-19 |
| EU-US Tariff Dispute eu_us_tariff_dispute | XME SPDR S&P Metals and Mining ETF | Long | 0.62 | Acknowledging the prior call (2026-08-19): I previously initiated a medium-term long on XME (confidence 0.60), citing European leadership de‑escalation of tariff risk plus non‑price evidence (CFTC copper net long +79,027; positive EU ZEW) that supports demand expectations and sector re‑rating. Critique of the prior view: the structural anchors used previously (COT positioning and EU sentiment) remain intact — the latest CFTC report (2026-08-11) shows net longs rising (week +4%) and EU macro sentiment has not reversed — so the original causal chain (demand/positioning → tighter metals sentiment → XME gains) is still valid. New evidence check: price has continued to validate the thesis (arc window XME +9.65% vs MXI +3.18%, excess +6.46%; 33d price_in excess = 3.22%; 33d etf_window cum +9.1% / max DD -7.18%), but per rules price is confirmatory only. Crucially, non‑price quantitative evidence remains supportive: CFTC copper net long = +79,027 contracts (100th percentile; OI up to 297,047; weekly +4%), which maps to the demand/positioning→sector rerating transmission. Contra evidence: extreme positioning (crowded COT), low options IV percentile (4% → cheap protection), recent net redemptions since 2026-08-13 of -$89M (-1.97% AUM) and accumulating short interest (+16.9%, DTC=2.67) raise reversal risk. Net assessment: the non‑price structural evidence has been reinforced and price action corroborates it, so I modestly strengthen the prior long (confidence 0.62) while flagging crowding and low IV as reasons to size/manage risk conservatively. | +10.8% | 2026-08-19 |
| US Debt Crisis us_debt_crisis | GLD SPDR Gold Shares | Long | 0.60 | Since the prior contested stance (2026-08-20), GLD’s price has partly priced the story (price_in excess = +8.94%), but multiple new, gold-specific non-price quantitative signals strengthen a bullish thesis: 1) CFTC COT net long rose to 141,648 contracts, week-over-week +3,986 (+3%), at the 100th percentile — indicating fresh speculative/hedge-side net longs; 2) GLD AUM saw net creations of $4,529M since 2026-08-10 (+2.94% AUM), 30d inflows +3.45% AUM — real-money demand; 3) 10Y real rate fell to 2.35% (20d −8bp), lowering the opportunity cost of holding non-yielding gold. Price motion is consistent: etf_window 9d cumulative +5.59% (max drawdown −1.71%) and arc excess +8.94%. The causal chain (us_debt_crisis → repricing of duration/credit risk → real rates & risk premium moves → institutional/speculative buying of gold/GLD) is supported by these independent non-price signals. I therefore take a medium-term long view on GLD, tempered by crowding indicators (IV percentile 96%, RSI 71, volume_flow 1.65x). | +5.6% | 2026-06-17 |
| Latin America Economy latin_america_economy | EEM iShares MSCI Emerging Markets ETF | Long | 0.58 | Maintain a mid‑term long view on EEM. Prior assessment (2026-08-18) strengthened the long view to confidence 0.62 based on two non-price supports: (1) fund flows (net creations/ins) into EEM (~+$480M; +1.56% AUM) and (2) a weak US retail sales print suggesting higher odds of marginal monetary ease that would favor EM. In this daily refresh those structural, non-price supports are intact: the net inflows remain (+$480M since 2026-08-14; 30-day +3.25% AUM) and VXEEM is low (26.54, z=-2.27), consistent with compressed EM risk premia and continued fund allocation into EM. Price has already reflected part of the thesis (arc window cumulative EEM +3.29%, EEM‑SPY excess +1.49%, max drawdown -7.0%), so some gains are realized and should not be double-counted as new evidence to increase conviction. Countervailing/mixed signals — ATM IV at 8th percentile (cheap protection), a relative skew difference vs SPY, short interest +3.8% (74.9M shares, days-to-cover 2.98), and Fed 4-week net liquidity change still negative but less so (-$125B) — argue against further confidence increases. Therefore keep long, but reduce confidence modestly to 0.58 to reflect partial price realization and crowding/complacency risk. Quantified anchors: price_in excess = +1.49%; etf_window cum = +3.29%; max drawdown = -7.0%; EEM net flows = +$480M (+1.56% AUM) since 2026-08-14; 30d flows = +3.25% AUM; VXEEM = 26.54 (z=-2.27); ATM IV percentile = 8%; FINRA short interest = 74,906,755 (+3.8%), days-to-cover = 2.98. | +1.9% | 2026-08-14 |
| Latin America Economy latin_america_economy | UUP Invesco DB US Dollar Index Bullish Fund | Short | 0.56 | Maintain the short USD (UUP) thesis: the earlier causal chain — a macro demand shock (US retail sales surprise on 8/14) plus fund withdrawals from the dollar ETF — still underpins a bearish USD view. The fresh snapshot shows persistent fund outflows (UUP AUM ~$0.4B; cumulative net outflows since 2026-08-03 ≈ -$66M, ~-18.7% AUM; 30d flow -21.13% AUM), which continues to signal capital leaving the ETF. However, confidence is lowered versus the prior view because the market has already priced a meaningful portion of the move (arc-window cumulative return -2.18%, max drawdown -2.52%, UUP -0.92% on 2026-08-19), and short-position crowding increased (FINRA short interest +22%, days-to-cover = 3.22), raising squeeze risk. Fed-funds implied path (~3.765%) and Treasury yields (2y=4.19%, 10y=4.71%) do not provide new hawkish support to overturn the thesis, but they do not strengthen it either. Options IV data are sparse (IV percentile 38%) and not reliable as an independent signal. Net: remain short, reduce conviction to reflect realized price move and increased crowding. | -4.3% | 2026-08-14 |
| BoJ Yen Normalization boj_yen_normalization | UUP Invesco DB US Dollar Index Bullish Fund | Short | 0.55 | Maintain a short on UUP (medium-term view: dollar down / JPY normalization). Non-price, structural evidence favors a weaker dollar: (1) CFTC shows large speculative net short JPY (≈ -67,971 contracts, report 2026-08-18), which creates a meaningful short-covering vulnerability should BOJ/official intervention or risk re-pricing occur — an outcome that would strengthen JPY and weigh on USD; (2) UUP demand is weakening (AUM $0.3B; 30‑day cumulative net outflow = -21.32% of AUM since 2026-08-03), indicating diminished liquidity/support. Price corroboration is present but secondary: 4‑day price_in excess = -0.63% (UUP -0.85% vs benchmark -0.23%), 54‑day window cumulative = +0.22% with max drawdown -2.52%. Historical analogs are directionally consistent (#401 6d excess -2.09%; #3418 5d excess -0.29%). Key risk is crowded UUP shorts (FINRA short interest +22%, days‑to‑cover=3.22) and imminent macro prints (US core PCE/GDP ~Aug 26) that could provoke short squeezes. Net: short retained, confidence raised to 0.55 on the arrival of independent non‑price evidence supporting the thesis; manage sizing ahead of near‑term macro events. | -0.9% | 2026-06-04 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | FXY Invesco CurrencyShares Japanese Yen Trust | Contested | 0.50 | Maintain contested. The core trade-off remains: CFTC positioning shows large speculative net short (previous −53,070 → current −67,971 contracts), which is a non‑price signal supporting a potential short‑squeeze (pro‑FXY), while macro/market rate signals (US−Japan 10Y spread ≈ +2.02pp and implied Fed‑funds path ~3.63%) continue to structurally favour a stronger USD (anti‑FXY). Price‑derived readings do not resolve the conflict: 52d price_in excess −2.85%; 52d etf_price_window cumulative +0.73% with max drawdown −2.49%; arc window cumulative FXY +0.35% vs SPY +3.82% → excess −3.47%. The triggering event (China‑Indonesia ministerial meeting) is diplomatic and contains no matching non‑price intervention signals (no official FX intervention, reserves move, or capital‑flow data). Therefore evidence remains conflicted; keep contested with confidence 0.50. | -2.9% | 2026-06-07 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | XLF Financial Select Sector SPDR Fund | Contested | 0.50 | Recap of prior: I previously labeled XLF contested (confidence 0.55 on 2026‑08‑21), citing non‑price signals (prediction markets, ETF redemptions, tightening Fed net liquidity) that pointed to downside pressure on banks but contrasted with strong price action. Current view remains contested. The gold rebound / gold‑ETF inflows reinforce the narrative “weaker macro → market lowers hike odds → compressed short‑end yields & NIM → banks under pressure,” but the evidence mix is ambiguous and not decisive. Non‑price support for downside: prediction‑market shows Fed‑hike yes_prob ≈ 0.45 (no extreme pivot), XLF has real ETF redemptions −$1,728M (≈ −3.07% AUM since 2026‑08‑13), and Fed net liquidity WALCL−TGA−RRP is $5,792B (4‑week Δ −$125B). Countervailing signals: price has materially outperformed (price_in excess 5.62%; etf_window cum 11.42%, max drawdown −2.25%), option IV is low (IV percentile 36%) and market breadth/technicals are supportive. Historical analog strength is weak (many event_opp entries flagged as first_signal). Conclusion: remain contested for the 1–3 month arc; do not enlarge a directional position now. Wait for the upcoming core PCE/GDP prints or clearer, sustained moves in non‑price indicators before committing. | +5.6% | 2026-05-14 |
| Metals Price & Demand metals_price_demand | GDX VanEck Gold Miners ETF | Long | 0.48 | Tilt long. Independent non-price demand/flow signals continue to support the causal chain “higher gold demand → re-rating of miners.” Key quantitative evidence: price_in (3d GDX excess) = +6.39% (GDX +5.92% vs SPY -0.47%), indicating concentrated short-term buying; more importantly non-price evidence includes (1) fund flows & volume: net creations since 2026-08-13 +$224M (+0.74% AUM), 30-day +1.53% AUM, 90-day +2.75% AUM, and elevated dollar volume ($3,328M vs baseline $1,815M, unusual ratio 1.53x) — direct quantification of ongoing demand; (2) macro real rates: FRED 10Y real rate = 2.35% (20d -4bp), easing the opportunity cost of gold holdings; (3) positioning & skew: CFTC speculative net long = 137,662 contracts and ATM IV percentile = 92%, showing positioning and risk premia changes. Offsetting concerns: etf_window cumulative return is large (+33.87%) with a max drawdown -10.16%, RSI=72.8 and elevated IV signal crowding/追高 risk. Net: non-price evidence strengthens the demand-repricing thesis so we increase conviction modestly (0.40 → 0.48) but keep tight risk controls and re-check after upcoming macro prints. | +33.4% | 2026-06-23 |
| BoE UK Gilt Cycle boe_uk_gilt_cycle | TLT iShares 20+ Year Treasury Bond ETF | Short | 0.45 | The RSM UK signal and corroborating macro data shift the tilt toward higher long-end yields and thus make a short stance on TLT appropriate over the 1–3 month arc. Causal chain: UK macro strength (e.g., gb_pmi_services = 52.8 vs 51.8, σ ≈ +0.5) raises the probability that the BoE remains less dovish / more hawkish → risk of gilt repricing higher → upward pressure on global long-term yields → downside pressure on long-duration Treasury exposure (TLT). Structural quant evidence: US long yields remain elevated (10Y = 4.69%, real 10Y = 2.35%, 2s10s = 0.50), CFTC treasury_30y net short = -361,383 (near 100th percentile) consistent with a market biased toward higher yields absent strong sustained buy-side interventions, and arc-window cumulative performance shows TLT -3.06% excess (TLT vs benchmark). ETF-level flows (30d net inflow +9.29% AUM) moderate but do not negate the macro-driven repricing risk; the previously-cited US Treasury buyback lacks quantified, persistent confirmation to outweigh the new macro signal. Therefore flip prior long to short with moderate-low confidence (0.45). | -4.0% | 2026-07-15 |
| BoE UK Gilt Cycle boe_uk_gilt_cycle | AGG iShares Core U.S. Aggregate Bond ETF | Long | 0.45 | I upgrade the medium‑term tilt toward AGG (direction = long) and raise conviction (0.45) because the trigger — a macro release indicating UK growth is expected to slow after a strong first half — is a direct non‑price signal that lowers the probability of further BoE hawkishness. Causally: weaker UK growth → lower BoE terminal/near‑term tightening odds → downward pressure on gilt yields → a reduction in global term premium/uncertainty → dampening of upward pressure on US long yields → supportive for AGG. This structural chain is reinforced by market non‑price signals supportive of duration: market‑implied Fed‑funds path remains tilted lower (ZQ implied ≈ 3.63%), and AGG has seen real net creations (30/90d net inflow +0.38% AUM), indicating persistent demand for aggregate duration. Price cross‑checks do not contradict this (arc window AGG +0.23% vs SPY +2.46%, price_in excess = -2.24%; 15d price_in excess = -0.97%; etf_window cum 0.09%, max DD -0.48%). Offsetting risks: elevated 10Y = 4.69% and term premium 0.8393%, and options market skews/IV (IV percentile 92%, skew +3.49pt) show continued demand for downside protection. Net: the new UK macro release weakens the gilt‑hawkish narrative that caused the prior downshift; combined with continued AGG inflows and market pricing of Fed easing, the evidence now modestly strengthens a duration‑friendly stance toward AGG. | -1.0% | 2026-08-01 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | IYR iShares U.S. Real Estate ETF | Long | 0.45 | Acknowledging the prior (2026-08-19): I previously kept a medium-term long on IYR (confidence 0.52), driven by the chain “weaker macro → marginal Fed dovishness → lower term rates → benefit to rate-sensitive assets,” supported by prediction-market signals and real ETF flows. Since then: (a) the historical reference (i.e. sustained macro weakening → dovish Fed) is less secure — recent 7-day PMIs are not weak (e.g., German manufacturing PMI 54.1, σ=+2.0), and prediction-market probability for a Fed hike in 2026 has risen from ~0.36 to 0.47 (manifold), weakening the dovish pivot thesis; (b) system liquidity continues to tighten (WALCL−TGA−RRP = $5,792B, 4-week Δ = -$125B), a persistent headwind for risk assets; (c) price and flow signals are mixed: IYR window cumulative +2.74% vs SPY +5.80% → excess = -3.06%, max drawdown -4.37% (relative underperformance), while real asset flows show modest inflows (since 2026-08-14 net +$16M ≈ +0.34% AUM; 30-day +0.54% AUM). The triggering news — a heavily funded AI/semiconductor startup poaching Nvidia talent — is an industry growth/capex signal (ai_compute_capex_race) that could, in theory, lift growth/inflation expectations and thus be hawkish for policy; but it is an idiosyncratic, small-sample signal and lacks direct, measurable Fed-policy changes. Net: I keep a medium-term long on IYR but reduce conviction (0.52 → 0.45, change_vs_prior=weaken). The weakening reflects the split non-price evidence (flows vs prediction-market + liquidity) and the fact that the new industry-capex narrative is an indirect and currently unquantified channel to Fed policy. Tactical note: maintain position size discipline and tighten before core PCE / FOMC windows. | -3.1% | 2026-06-03 |
| ECB Eurozone Rate Cycle ecb_eurozone_rate_cycle | KBE SPDR S&P Bank ETF | Short | 0.45 | Updated thesis (vs prior: contested, confidence 0.30, where macro weakness favored dovish central banks but flows/short crowding argued against short execution). I flip to short. Rationale: 1) The triggering event (event_opp #3881) describes a Treasury rally and market pricing-in of easing inflation — a market-level dovish rate impulse consistent with the transmission: central bank dovish surprise → lower term rates → compression of bank NIM. 2) Non-price position/flow evidence has shifted against KBE: since 2026-08-10 KBE has seen real creation/redemption net outflows of -$151M (-8.38% AUM, 5 events as of 2026-08-17), indicating actual capital leaving the ETF and reducing prior execution risk from inflows. 3) Price-relative measures corroborate this structural view: arc cumulative KBE +8.61% vs XLF +10.53% → excess -1.92%, and longer price_in_score = -2.93% (KBE 6.99% vs XLF 9.92%), consistent with regionals underperforming broader financials in a falling-rate environment. Offsetting risk: FINRA short interest remains elevated (15,145,952 sh, days-to-cover = 11.2; 7-period short accumulation +265.2%), so squeeze risk is real; adopt a modest-sized short with event-driven risk management, and re-evaluate after major inflation prints / ECB commentary. | +0.4% | 2026-06-11 |
| US Debt Crisis us_debt_crisis | KRE SPDR S&P Regional Banking ETF | Short | 0.45 | Maintain a mid-term short on KRE (regional banks). Rationale: the structural channel remains intact — NY Fed data (event_opp #3835) shows 90+ day credit-card delinquencies rising, which pressures regional bank asset quality and margins. This refresh provides independent non-price quantitative reinforcement: true creation/redemption flows show net outflows of −$427M since 2026-08-14 (−10.51% of AUM; AUM = $4.1B as of 2026-08-20), 30-day cumulative flows −8.69% AUM (19-day sample), 90-day −17.45% AUM — indicating real capital flight from the ETF that amplifies selling pressure. Price corroborates this: price_in excess = −3.92% (KRE −1.47% vs XLF +2.44%, 2026-07-21→2026-08-21), etf_window cumulative −1.47%, max drawdown −4.13%; per rules, price is confirmatory but not the primary directional driver. Other quantitative context: option IV is low (IV percentile 14%), and Fed-related liquidity (WALCL−TGA−RRP) is $5,792B with a 4-week change of −$125B, supporting concerns about liquidity/credit sensitivity. Conclusion: the real outflow evidence independently strengthens the short thesis; confidence is nudged up within the daily-refresh cap to 0.45. | -3.8% | 2026-07-21 |
| US-Iran 2026 War us_iran_2026_war | UNG United States Natural Gas Fund | Contested | 0.44 | Remain contested. Structural bearish, non-price evidence persists but is not decisive: CFTC COT (report 2026-08-18) net = -100,080 contracts (4-period series: -105,826 → -126,933 → -110,630 → -100,080; weekly +10,550; OI up to 1,727,378), indicating a material speculative short position; Henry Hub forward curve is in contango (near-month 2.773 vs 12-month 3.20 → slope -13.344%), implying roll drag and no clear storage tightness — both are direct non-price channels that argue against UNG. Offsetting that, demand/positioning show buy-side support: UNG authorized participant flows since 2026-08-03 +$52M (+10.33% AUM), 30-day cumulative inflows +11.99% AUM (crowded/price-chasing risk); option IV is low (historical percentile 12%), making protection cheap; prediction markets price a high probability of short-term de-escalation (polymarket ceasefire through 8/31 yes_prob=0.86), reducing near-term risk-premium upside. Price has partially moved (arc window excess = -5.49%, max drawdown -20.54%, latest close 9.99) and therefore has already baked in some downside. Because non-price signals conflict (COT + contango vs flows + low IV + prediction markets) and no new conclusive non-price evidence arrived to break the tie, keep the contested view and modestly lower conviction to reflect partial price realization and elevated uncertainty. | -9.3% | 2026-05-14 |
| AI Compute Capex Race ai_compute_capex_race | SOXX iShares Semiconductor ETF | Contested | 0.42 | Maintain contested. Prior: I downgraded from short (conf 0.66) to contested (conf 0.45) because the prior short thesis—"sustained redemptions → valuation pressure"—was undercut by a reversal in fund flows (net creation +$1,543M since 2026-08-14) and stronger industry signals (DRAM spot up), while options and short-positioning still signaled defensive/bearish behavior, yielding conflicted evidence. Current refresh: FRESH SNAPSHOT provides no new decisive non-price evidence to resolve that conflict. Quant facts: arc-window SOXX cumulative -2.39%, SOXX-QQQ excess -2.41%, window max drawdown -29.01%, latest close 520.05 (2026-08-21). Non-price signals remain split—flows (net creation +$1,543M / +3.64% AUM; 30d flow -0.60% AUM) and DRAM spot (TrendForce DDR5_16Gb +13.85%) support the bullish capex demand chain; options/vol and positioning (ATM IV call 40.53% / put 42.54%, put-call skew +2.01pt; skew vs QQQ +3.97pt; IV percentile 0%) and short interest (12,611,021 shares, +1.7%, days-to-cover 1.05) support defensive/hedging pressure. Per rules, direction must rest on non-price quant signals; they conflict and no fresh event reconciles them. So remain contested and slightly lower confidence to reflect time decay and partial price realization (but price alone does not decide direction). | -6.0% | 2026-07-20 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | VNQ Vanguard Real Estate ETF | Short | 0.42 | Acknowledging the prior view: On 2026-08-19 we stayed modestly long VNQ with confidence 0.35 but had downgraded conviction because the market-implied probability of a 2026 Fed hike had risen and Fed net liquidity was falling. Re-assessing now: (a) The prior analogue of a dovish pivot benefiting REITs has weakened. The prediction market probability that the Fed hikes in 2026 has increased to 0.47 (via Manifold, 2026-08-21), further diverging from a dovish setup. Fed net liquidity stands at $5,792B with a 4-week change of -$125B (as of 2026-08-19), still contracting. (b) Tape check: VNQ is up +1.74% over the arc window but underperforms SPY by -0.94% with a max drawdown of -4.53%. Price has not validated the prior long (used only as corroboration, not as a directional driver). (c) The new event (AI startup talent move) has no causal link to the Fed path or REIT financing and is neutral for this arc. Non-price quantitative drivers now tilt hawkish: a 0.47 yes_prob implies non-trivial risk of another hike; persistent liquidity drainage limits duration relief and keeps financing/cap rates pressured for REITs. VNQ ATM options IV sits at the 8th percentile (low), which is not the kind of elevated fear that would support a contrarian equity-bullish read. Netting this, the ‘dovish pivot’ blueprint lacks fresh, asset-specific confirmation while the ‘hawkish/duration pain’ blueprint gains support. We therefore flip to a cautious short over 1–3 months, with 8/26 core PCE as a key checkpoint that could reverse this view if it meaningfully undershoots. | -0.9% | 2026-07-01 |
| Latin America Economy latin_america_economy | EWZ iShares MSCI Brazil ETF | Short | 0.42 | Maintain a medium-term short on EWZ, with slightly higher conviction. The primary causal chain is demand deterioration: real fund flows show continued withdrawals (30-day net redemption = -10.27% of AUM; since 2026-08-14 net outflow -$458M, ≈ -5.82% AUM), undermining ETF demand. Price signals corroborate relative weakness—30-day price_in excess = -4.99%; over the arc window (2026-07-13 → 2026-08-14) EWZ cumulative -4.13% vs EEM +3.27% => excess -7.40%; 10-day etf_price_window cumulative -0.37% with max drawdown -4.23%. These metrics support the chain: outflows → demand/liquidity pressure → relative underperformance → medium-term downside risk. Offsetting evidence includes lower reported short interest (33,328,766 shares, -21.4% vs prior, days-to-cover=2.01) and muted EM volatility (VXEEM=27.05, z≈-2), which raise the prospect of short-covering or transient rebounds but do not negate the sustained outflow signal. Therefore keep short and modestly strengthen conviction, while monitoring for abrupt short-covering or volatility regime shifts. | -3.1% | 2026-07-13 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | TLT iShares 20+ Year Treasury Bond ETF | Contested | 0.40 | Maintain a contested view: non-price evidence remains conflicted. Pro-long-duration signals include meaningful TLT creation/redemption inflows—+$512M since 2026-08-17 (+1.11% AUM) and +9.29% AUM over 30 days—indicating investor demand. Opposing, direct counter-signals persist: CFTC treasury_30y net = −361,383 contracts (100th percentile vs recent range, implying crowded speculative short positioning that increases vulnerability to higher yields), market-implied terminal Fed funds ~3.735% (no clear market pricing of imminent cuts), 10y nominal/real yields remain elevated (10y ≈ 4.69%, real ≈ 2.35%), and MOVE ≈ 73.4 (moderate volatility). Price-derived checks: arc window excess = −3.43%; etf_price_window (9d) cumulative −0.17%, max drawdown −1.50%, latest close 82.05; option IV ~40th pctile. Today's news (chip startup hiring) is not a causal, on-the-book non-price driver for the Fed/rates arc, so it provides no new decisive evidence. Therefore keep contested and wait for a clear, arc-specific non-price trigger (substantive COT rebalancing, deteriorating auction demand, or materially lower implied Fed path) to pick a direction. | -10.1% | 2026-07-18 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | PFF iShares Preferred and Income Securities ETF | Contested | 0.40 | Acknowledging the prior call: On 2026-08-23 I moved PFF to contested (confidence 0.40) because the trigger was a startup fundraising headline, which is non-directional per our rule that financing is not a rate/credit signal; meanwhile non-price data remained mixed: the implied Fed funds path rose to 3.735% (vs ~3.63% prior, a marginally more hawkish tilt) and Fed net liquidity fell by ~$125B over 4 weeks, yet credit and plumbing were calm (HY OAS 2.75%, SOFR−IORB -2bp, MOVE 73.4). Re-examination: (a) Historical analogs still apply but are not strongly engaged—‘hawkish surprise → duration pain’ is a valid template for preferreds, but the front-end path remains below current policy (cuts still priced) and rates vol is subdued, so no clean hawkish trigger; the ‘dovish pivot → premium for rate-sensitive assets’ analog also lacks reinforcement as ZQ ticked higher and HY OAS widened 4bp over 5d. (b) Tape/flows lean against bulls but are secondary: arc excess return is -2.71% (PFF -0.03% vs SPY +2.68%), window cumulative -0.03%, max drawdown -2.08%; 30d net outflow -1.78% AUM; technicals show MA20<MA50 and RSI14=46.5. (c) The new ‘fundraising’ event has no direct transmission channel to this arc and is neutral context. Direction: keep contested. Non-price signals remain cross-cutting—front-end still prices easing (ZQ below current, prediction market puts 2026 hikes at 0.47), while the long end and credit are headwinds (10Y 4.69%, term premium ~0.84%, HY OAS 2.75% with +4bp over 5d). Calm volatility and stable funding do not justify a committed stance. Price/flow weakness only shows it’s not being bid yet and cannot decide direction on its own. With core PCE/GDP imminent, neutrality is preferable. | -2.7% | 2026-07-01 |
| ECB Eurozone Rate Cycle ecb_eurozone_rate_cycle | PFF iShares Preferred and Income Securities ETF | Contested | 0.40 | Keep contested (mixed/neutral). The prior split narrative still holds: market pricing offers some marginal dovish support for rate-sensitive assets, but funding/positioning flows remain a clear counterweight that prevents a unilateral long view. Price-derived metrics show continued relative underperformance (arc-window excess = -2.88%: PFF -0.07% vs SPY +2.81), etf_window cumulative -0.07% with max drawdown -2.08% and last close 30.46 — indicating the lag persists. Independent (non-price) signals remain mixed: market-implied Fed funds 3.735%, 10Y=4.69% (term premium ≈0.8393%) and MOVE=73.4 are consistent with a non-hawkish rate backdrop, but there is no ECB-specific dovish read. Conversely, flows/positioning are negative for PFF (30d net flows -1.72% AUM; net outflows -$60M since 2026-08-14; AUM $13.1B) and short interest is elevated (FINRA shorts 5,049,497 sh, +11.3%, days-to-cover 1.85). Therefore the arc remains contested; given continued outflows and partial price realization/time decay I lower conviction slightly versus prior. | -2.9% | 2026-07-02 |
| ECB Eurozone Rate Cycle ecb_eurozone_rate_cycle | QQQ Invesco QQQ Trust (Nasdaq-100) | Contested | 0.40 | Remain 'contested'. Prior (2026-08-19) flagged a collision: ECB dovish pivot would favor rate-sensitive growth/tech (QQQ), while holdings and flow/position crowding created a constraining opposite force. Update summary: price-derived metrics still show structural underperformance—QQQ vs SPY excess = -4.30% (price_in), and the 9-day etf_price_window cumulative = -1.38% with max drawdown -2.89% (latest close 713.44 as of 2026-08-21)—so the medium-term disadvantage remains but short-term rebounds have occurred. Crucially, there is no new ECB-side non-price dovish signal; instead, euro-area macro is firmer (Germany manufacturing PMI = 54.1, surprise σ=2.0), which reduces the likelihood of an ECB dovish pivot and therefore weakens the causal chain 'ECB dovish → rate re-pricing → growth/tech rally'. CFTC (Nasdaq) remains net-short though it has contracted (net = -61,771 contracts; week +27,354; percentile 89), QQQ shows continued inflows (30d +2.36% AUM; 90d +1.60%), and option/vol metrics show complacency (IV percentile 16%; VXN = 23.26 z=-1.04; call IV 19.96% vs put IV 18.0%). Net: the fresh non-price evidence tilts against the immediate dovish-trigger narrative and raises crowding/squeeze risk; lower conviction vs prior—keep contested but weaken (confidence 0.40). | -4.3% | 2026-06-11 |
| US-China Tech Decoupling us_china_tech_decoupling | SOXX iShares Semiconductor ETF | Contested | 0.40 | Prior (2026-08-20 11:49) was contested (conf=0.45): I emphasized the structural chain export_controls → weaker demand/market share for firms exposed to China, with mixed evidence—short-interest and longer-term outflows argued for downside, while DRAM spot and volatility signs argued for possible rebound. I left the arc contested and lowered confidence. This daily refresh remains contested. Fresh snapshot supplies opposing non-price quantitative signals: on the positive side DRAM spot prices are strongly up (aggregate +9.16%, e.g. DDR5_16Gb +13.85%) and SOXX has seen net creation inflows since 2026-08-14 of +$1,543M (+3.64% AUM), indicating recent capital return; on the negative/risks side short-interest remains elevated (FINRA short_shares=12,611,021, +1.7% vs prior; days-to-cover=1.05; 7-period short accumulation +70.8%) and volatility index VXN is low (22.04, z=-1.46), increasing squeeze risk. Price-in has already partially realized the downside: arc-window SOXX -5.93% (SOXX vs QQQ excess -4.98%), window max drawdown -16.29%. Because non-price signals conflict and price has already reflected part of the bearish view, I retain a contested call and reduce conviction from 0.45 to 0.40. | -7.8% | 2026-07-15 |
| US-Iran 2026 War us_iran_2026_war | IAU iShares Gold Trust | Long | 0.38 | My prior contested view rested on the offset: large speculative COT longs vs rising real rates and high de‑escalation probability in prediction markets. Since then the balance of non‑price evidence has shifted. CFTC net speculative longs rose to +141,648 contracts (reporting period 2026‑08‑18), with open interest increasing to 406,260 — a sign of new, not merely rolling, long positions. Real 10y real yields eased to 2.35% (20d -8bp), reducing the opportunity cost of holding non‑yielding gold. ETF-level activity shows net creations since 2026‑08‑17 of +$315M (+0.46% AUM), and the arc window shows IAU excess return +6.65% (52d). These three non‑price signals form a consistent causal chain: elevated geopolitical/risk premium → speculative/ETF accumulation (COT↑, inflows) → price rerating, with declining real yields making the position structurally easier to hold. Prediction markets remain a counterargument (polymarket: ceasefire through 8/31 yes_prob=0.92), and option IV is elevated (IV historical percentile 79%), implying risk‑premium/ crowding. On balance I flip to a tactical long for the 1–3 month arc (confidence 0.38) but keep position sizing conservative and monitor COT flow, real yields, and prediction‑market signals; if COT/OI roll over or real yields spike again, revert to contested/flat quickly. | +3.1% | 2026-06-06 |
| BoJ Yen Normalization boj_yen_normalization | IYR iShares U.S. Real Estate ETF | Contested | 0.38 | Maintain contested on IYR (no clear long/short call) but with reduced confidence. The arc hypothesis is: BOJ/yen normalization → wider JGB‑US or higher US yields → pressure on rate‑sensitive equities (IYR). The refresh shows mixed price vs non‑price signals: price_in over the arc window: IYR cumulative +4.08% vs SPY +3.34% (excess +0.74%), window max drawdown −4.37%, latest close 104.73 (2026‑08‑21) — indicating prior expectations have been at least partially priced. There is no new non‑price, “on‑the‑chain” evidence (e.g., a clear move in JGB‑US spreads or US yields) that would strengthen the causal link. Independent non‑price data: net creations since 2026‑08‑14 +$16M (+0.34% AUM), 30d +0.54%, 90d +2.57% (AUM $4.6B), and FINRA short interest down −13.8% to 9,696,388 shares (days‑to‑cover 1.86) — these give gentle funding/position support for IYR. Offsetting constraints: Fed net liquidity 4w −$125B (WALCL−TGA−RRP $5,792B), option surface with put IV 15.36% vs call IV 13.82% (skew vs SPY +2.70pt) and IV at 0% vs recent 9‑day history (low), and a calm VIX term structure (VIX/VIX3M 0.818). Net: flows and short covering persist as mild support but no new “pairwise” non‑price evidence to turn the BOJ→US yield→IYR transmission into a decisive signal; price has already reflected part of the thesis → lower conviction (from prior 0.45 to 0.38). | +0.7% | 2026-06-17 |
| BoJ Yen Normalization boj_yen_normalization | FXY Invesco CurrencyShares Japanese Yen Trust | Long | 0.38 | Maintain a long bias on FXY: the causal chain remains intervention (BOJ/MOF or coordinated support) → forced short-covering of crowded speculative JPY shorts → JPY strength → FXY appreciation. Non-price positioning evidence still underpins this thesis: CFTC speculative positioning remains materially net short (net short = -53,070 contracts as of 2026-08-11; total OI = 391,874; 4-period series -96,185 → -101,990 → -60,825 → -53,070; weekly change +7,755 / +13%; percentile = 100), which creates mechanical squeeze risk if authorities act or signaling intensifies. Supporting context: FXY 30d net flows +1.23% AUM (since 2026-07-10) and ATM IV at a low historical percentile (22%) indicate limited paid downside protection. Quant cross-check: arc price_in excess = -2.50% (FXY +0.66% vs SPY +3.16%), and the 8-day etf_window shows cumulative +0.69% with max drawdown -0.26% — price has partially realized the thesis but not fully. Therefore remain long but reduce confidence from prior 0.48 to 0.38 because part of the expected move is already priced, the arc has aged without fresh independent non-price confirms (no new COT regime shift or new official intervention announcement), and position should be managed for follow-on policy triggers. | -2.9% | 2026-06-07 |
| ECB Eurozone Rate Cycle ecb_eurozone_rate_cycle | FXI iShares China Large-Cap ETF | Long | 0.37 | Maintain a mid-term long bias on FXI. The thesis remains rooted in non-price, structural signals: (1) continued official Chinese allocation into FX and gold (July official gold accumulation +640k oz; FX reserves +$2.5bn to $3,418.8bn) provides structural support for the RMB and Chinese assets, reducing systemic downside and supporting relative equity valuations; (2) U.S. macro softness previously cited remains a moderating force for further Fed tightening and the USD. Price metrics are supportive but secondary: 30d price_in excess (FXI vs EEM) = +3.17%, FXI 30d cumulative = +7.24% with a max drawdown -4.55% and latest close 35.86; arc-window FXI-EEM excess remains positive (+5.20%). Crowd/flow constraints (short interest up +72.4%, days-to-cover = 3.25; 90d flows -1.09% AUM) and relatively low IV (29th percentile) argue for measured sizing. The July official gold/FX-reserve release confirms an already-identified transmission channel rather than providing a new independent pivot, so I keep the prior long stance with only a modest confidence adjustment to 0.37. | +5.0% | 2026-07-12 |
| China Internet Platforms — Earnings/Users/Consumption china_internet_platforms | KWEB KraneShares CSI China Internet ETF | Contested | 0.36 | Remain contested: the index restructuring still provides a plausible structural bullish channel (expansion → potential passive inflows → Hong Kong tech benefit), but without quantifiable, arc-specific evidence of passive inflows the causal chain is unconfirmed. Since the prior note, price and independent non-price signals have not validated an expansion-driven rally: price_in_excess = -2.06% (KWEB cumulative +5.63% vs FXI +7.69%), etf_window max drawdown = -8.45%, consistent with historical analog #559’s downside, indicating the market has not priced in a clear expansion-led lift. Non-price indicators do not support a surge of passive inflows; instead, short interest has continued to accumulate (41,541,170 shares, +16.4% over 7 periods, days-to-cover=1.93), options IV is at a low 9th percentile (complacency), and Fed net liquidity has contracted by ~$125B over 4 weeks — none of which corroborate a near-term passive inflow story. Net: remain contested, but confidence is lowered because price partially realized the absence of supportive flow evidence and crowding/liquidity signals are cautionary. | +3.1% | 2026-06-18 |
| US-China Tech Decoupling us_china_tech_decoupling | LIT Global X Lithium & Battery Tech ETF | Long | 0.36 | Switch to long (medium term, 1–3 months). The triggering event provides direct, non-price quantitative evidence tied to lithium demand: per event_opp #4119 (Han Xin, 2026-08-18) China’s EV monthly penetration exceeded 60% for the first time and cumulative penetration exceeded 50%; charging points reached 23.057 million (+43.2% YoY) with 98.61% county coverage; EV exports >50% share for two months and annual exports may exceed 4.0M units. Causation chain: event → higher EV sales & charging infrastructure → higher battery demand → stronger lithium demand → positive for LIT. Price action has already shown stress (arc window LIT −11.46% vs MXI −2.51%, excess −8.95%; etf_price_window cum −10.8%, max drawdown −20.17%, latest close 74.02), so some negative information is priced in. Non-price corroboration includes options-side call-crowding (put_call_vol_ratio 0.28; call_vol 67 vs put_vol 19) and commodity positioning (CFTC copper net long 79,027) supportive of commodity/industrial metals sentiment. Offsetting risks: tighter liquidity and macro softness (Fed 4-week net liquidity change −$191B; US retail sales −0.6% vs +0.1% exp, surprise ≈ −2.8σ) and recent outflows (LIT 30d flows ≈ −2.14% AUM). On balance, the event provides the required 'on-path' non-price demand signal for a medium-term long on LIT, but with moderate-low conviction (confidence=0.36) and position sizing discipline; monitor EV shipment continuity and lithium supply/inventory or capex signals for re-assessment. | -10.4% | 2026-06-16 |
| US-Iran 2026 War us_iran_2026_war | ITA iShares US Aerospace & Defense ETF | Contested | 0.35 | Event note: today’s reports that the U.S. might “indefinitely” rotate a blockade add theoretical weight to the chain (sustained blockade → higher ship operational tempo → increased mid‑term maintenance/logistics/parts demand → positive for ITA constituents). However, the report is narrative and lacks direct, measurable ‘on‑the‑books’ evidence (contracts, budget appropriations, procurement amounts). Quant metrics are mixed: price-derived signals show the arc has materially repriced ITA (arc_window ITA +17.08% / SPY +6.04%, excess +11.05%; price_in_score 4.41%), while the short 9‑day price window is weak (etf_price_window 9d cum -5.78%, max drawdown -6.27%). Non‑price signals conflict: prediction markets price a sustained ceasefire (polymarket ‘ceasefire through Aug 31’ yes_prob=0.92, vol=$709,279), which argues against a durable demand shock, whereas positioning/flows are mixed (FINRA short interest 1,550,216 shares, +51.5%, days‑to‑cover=2.12 shows active shorts; ITA 30‑day net inflow +1.84% AUM shows buying). Options IV is low (6th percentile), implying cheap tail protection. Net: the structural blockade→demand narrative remains plausible and partially price‑reflected, but absent direct non‑price procurement evidence and with strong prediction‑market odds for de‑escalation plus short activity and recent price weakness, I keep the view contested but reduce confidence (0.35). Remain low‑leverage/observe until clear non‑price confirmation (contracts/budget/prediction‑market shift) or material positioning changes occur. | +4.4% | 2026-07-21 |
| China Rare Earth Export Restriction china_rare_earth_export_restriction | SOXX iShares Semiconductor ETF | Long | 0.35 | The prior view (contested) was defensible because there was previously no direct, "on-path" export-control evidence. On 2026-08-21 China reportedly eased curbs on Nvidia H200 shipments (event_opp #4161), which is a direct policy inflection on semiconductor export controls and reduces downside tail risk to chip sales into China. Non-price quantitative support for a bullish tilt: (1) SOXX net creations/redemptions since 2026-08-17 = +$673M (+1.61% AUM), indicating real-money buying; (2) DRAM spot trend is up (spot main +9.16%; DDR5_16Gb +13.85%), supporting semiconductor demand; (3) options/positioning show heavy put protection (put_call_vol_ratio 5.87; put_call_oi_ratio 3.01; put_vol 1871 vs call_vol 319), which could unwind or be squeezed if policy easing persists. Price measures are supportive as corroboration but not decisive: arc window (2026-07-27→2026-08-21) SOXX cumulative +0.74% vs QQQ +4.59% → SOXX-QQQ excess -3.85%; window max drawdown -9.92%; latest close 520.05. Headwinds: concentrated short interest accumulation (near-term +70.8% over 7 periods; days-to-cover=1.05) and earlier longer-window outflows (30d -2.17% AUM; 90d -10.44% AUM) reduce conviction. Net: because today’s event is a direct policy easing that maps to the arc’s causal chain and is backed by fund flows and industry data, I flip from contested to a tactical long on SOXX for 1–3 months, at modest confidence (0.35). | -2.9% | 2026-07-26 |
| Trump 2026 Tariff Policy trump_tariff_2026_policy | MXI iShares Global Materials ETF | Long | 0.35 | Prior assessment judged the arc contested due to lack of tariff-specific non-price evidence. The Iran attack (heightened risk to the Strait of Hormuz) provides event-specific, non-price commodity signals: speculative copper net long = 79,027 and oil backwardation slope_pct = 14.62% (see event #3985/#3261). Causal chain: Middle East conflict → increased shipping/oil supply disruption risk → higher commodity prices and speculative flows → supports ETFs sensitive to metals/EM/resource beta such as MXI. Short-term analogs support a positive MXI response (event #3261: MXI 5d excess = +4.09%, flagged=True; price_in_score 43d excess = +3.63% (MXI 6.5% vs SPY 2.87%)). Offsetting risks: across the full arc window MXI underperformed (2026-06-22→2026-08-17: MXI +1.93% vs SPY +3.80% → excess -1.87%), etf_window max drawdown = -5.87%, small AUM ($0.4B), rapid short accumulation (FINRA short_shares=39,487, +80.7%, days-to-cover=2.98) and rising narrative heat (crowding). Conclusion: flip from contested to a cautious long on MXI for a 1–3 month horizon with moderate-low confidence; keep position sizing conservative and watch for direct arc-specific data (USTR announcements, EIA/OPEC supply confirmations) to re-evaluate. | +3.6% | 2026-06-20 |
| US-China Tech Decoupling us_china_tech_decoupling | BNO United States Brent Oil Fund LP | Contested | 0.35 | Maintain the prior contested stance. The prior view (2026-08-21 12:10) held that a sovereign-sanction / supply-tightness → near-month premium (backwardation) → BNO bullish chain exists but that much of this was already priced in and crowding risks warranted a contested posture. Today's daily refresh brings no new event to overturn that chain: non-price evidence still partially supports supply tightness (CFTC COT net long +87,479 contracts, week +7,563/+10%, at the 58th percentile; WTI near-month vs 12-month curve shows strong backwardation, ~17.76% annualized), which are direct, arc-relevant signals for a substitute/near-month premium. At the same time, crowding / price-in signals remain strong: arc-window excess return +21.9% (BNO cum +24.77% vs SPY +2.87%), ETF window max drawdown −15.21%, options IV at 12th percentile with 5d IV −3.11pt (implied vol falling/complacency), and FINRA short interest +48.4% with days-to-cover=1 (crowded positions/short-squeeze dynamics). Therefore, while COT and the curve continue to support the supply-tightness narrative, there is insufficient independent non-price evidence to confidently strengthen the bullish call; price action has already reflected much of the move—so we keep the contested view and hold confidence unchanged. | +21.9% | 2026-06-20 |
| ECB Eurozone Rate Cycle ecb_eurozone_rate_cycle | VGIT Vanguard Intermediate-Term Treasury ETF | Contested | 0.35 | Keep contested (no trade). The prior view that ECB/eurozone rate-cycle signals were mutually offset remains valid after the fresh snapshot: price has already delivered much of the arc move (price_in excess = +3.60%; VGIT arc-window cumulative = -0.97%; window max_drawdown = -1.59%), reducing available relative alpha and arguing for lower conviction. Independent, non-price signals are mixed: eurozone manufacturing PMIs came in notably hot (EU manufacturing 52.8 vs exp 51.8, σ≈1.2; Germany manufacturing 54.1 vs exp 52.1, σ≈2.0) and CFTC 5y shows persistent net-short positioning (treasury_5y net = -2,169,814 contracts; w/w change -22,070, -1%), both leaning toward duration pain; offsetting that, the market-implied Fed funds path fell to 3.63% (vs ~3.725% previously), and MOVE (73.4) plus strong auction bid-to-cover (~2.67) suggest calmer rates/liquidity—so evidence remains conflicted. Because directional inference must rest on non-price quantitative signals and they do not align into a single, decisive story, the arc remains contested and conviction is trimmed from 0.40 → 0.35. | -4.8% | 2026-06-11 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | SPY SPDR S&P 500 ETF Trust | Contested | 0.34 | Remain contested. The prior call (2026-08-21) was reasonable: non-price quantitative signals remain mixed — recent macro weakness (e.g., retail sales 2026-08-14 MoM = -0.6% surprise) and prediction markets (manifold fed-hike yes_prob rising from 0.42 to 0.47) support a dovish repricing → equity-friendly transmission, while opposing non-price signals (CFTC COT net = -281,402 contracts at the 98th percentile; Fed net liquidity WALCL−TGA−RRP = $5,792B, 4-week change -$125B; and sizable SPY net inflows: 30-day +1.45% AUM) point to crowding/liquidity withdrawal risks. Price cross-check: arc window cumulative SPY excess = +6.64% (2026-05-04 → 2026-08-21); 9-day ETF window cum = -0.63% with max drawdown -1.96% — short-term pullback within an intermediate positive trend. The current event (a large-funded semiconductor startup poaching Nvidia talent) is an industry/company story and does not provide a new, arc-relevant non-price quantitative trigger for Fed rate-path repricing. Therefore the stance remains contested until a clear non-price policy/ macro inflection arrives. | +0.0% | 2026-06-03 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | XLRE Real Estate Select Sector SPDR Fund | Long | 0.33 | Maintain a mid-term (1–3 month) long on XLRE. The arc-level structural thesis — a marginally dovish Fed or lower odds of further hikes leading to lower discount rates and a relative tailwind for rate-sensitive sectors — still stands. My prior assessment (2026-08-23 10:32:40) reached the same directional conclusion but noted ETF-specific counter-evidence. Re-evaluating adversarially: price-derived metrics over the arc show XLRE cumulative +2.04% vs SPY +2.68% (excess -0.64%) with max drawdown -4.19%, i.e., no decisive excess outperformance. Crucially, ETF-specific non-price evidence has become slightly more bearish: net creation/redemption flows since 2026-08-13 are -$172M (~-2.03% AUM), 30-day flows -0.76% AUM, and FINRA short interest is 7,454,300 (+4.6%, days-to-cover 1.56). Macro/flow signals remain ambiguous — prediction markets show ~47% probability of another Fed hike (not a strong hawkish signal), and system liquidity (WALCL−TGA−RRP = $5,792B, 4-week -$125B) has softened but not decisively tightened policy odds. Options IV is at low historical percentile (call IV 14.36%, put IV 10.3%, IV pctile 0%), implying market complacency rather than stress. The new event (#4165, a semiconductor startup poaching talent) is not causally linked to Fed policy or RE discount rates and therefore is not a valid non-price driver to change direction. Net: keep long but reduce confidence (from 0.40 → 0.33) and await the core PCE/GDP prints and any direct Fed communication before increasing exposure. | -0.6% | 2026-07-01 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | VOO Vanguard S&P 500 ETF | Long | 0.33 | Maintain a medium-term constructive stance on VOO but with reduced confidence. Price evidence still shows modest excess (price_in_excess = +1.49%, window cum = +1.49%, window max drawdown = -4.17%, latest close = 703.71), indicating resilience in risk assets, but price alone is not decisive. Non-price signals are mixed: CFTC COT positioning is extremely crowded net-short (net = -281,402 contracts, 98th percentile), which supports a technical squeeze/short-covering tail to the upside; by contrast, predictive markets imply a higher chance of further Fed tightening (manifold yes_prob = 0.47 vs ~0.41 previously) and system liquidity continues to decline (WALCL−TGA−RRP = $5,792B, 4w -$125B), both substantive headwinds to a dovish Fed pivot. The triggering news — a $2bn+ fundraising chip startup poaching Nvidia talent — maps to an ai_compute_capex_race narrative (possible incremental positive for tech/semiconductor capex expectations) but does not provide a direct, non-price causal channel to change the Fed path. Net: keep long exposure for now (price resilience + crowded positioning), but lower conviction (0.33) and wait for upcoming macro prints (core PCE, GDP) or Fed communications before increasing exposure. | -0.0% | 2026-06-03 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | EEM iShares MSCI Emerging Markets ETF | Contested | 0.33 | Recap of prior view: On 2026-08-18 I had maintained a modestly bullish stance on EEM (confidence=0.48, strengthen) based on non-price signals that suggested a softer Fed path (manifold ~0.36, weak US retail sales) plus institutional allocation flows (net creations since 8/11 ~+1.55% AUM, 30d inflows ~+3.23% AUM) and an event tagged as central-bank/FX-intervention which together supported the transmission “slower policy → dollar/term premium relief → EM outperformance.” Critique of that prior: it leaned on the assumption that policy-side (central bank) intervention would materialize to lower FX/market tail risk; if that policy channel weakens, the bullish chain is fragile. Current assessment and why it changes: the triggered item today is a China–Indonesia ministerial (2+2 diplomatic/defense) meeting — a diplomatic/strategic event, not an actual central-bank FX intervention. Therefore it does not serve as direct, non-price evidence of authorities stepping in to stabilize FX or widen liquidity. Meanwhile several key non-price, quant signals have moved against the prior dovish story: prediction-market probability of a Fed hike rose to 0.45 (from ~0.36 previously), arguing for a higher chance of tighter policy; system liquidity metrics remain contractionary (WALCL-based 4wk Δ ≈ -$125B); flows into EEM remain positive (recent net creations +1.56% AUM; 30d +3.25% AUM) but those are price-reflective/allocation signals and cannot alone set direction. Other structural observations: VXEEM is low (27.05, z=-2.06) and short interest has increased (74,906,755 shares, days-to-cover=2.98), indicating positioning crowding and potential squeeze dynamics. Net result: the policy/flow signals are conflicted; because the policy-side non-price evidence that previously supported the bullish chain has weakened and the event is not a direct intervention, I downgrade confidence and move the assessment to contested (weaken). Tactical implication (1–3 months): maintain caution — if continuing to hold a long tilt, materially reduce new-entry size and wait for clear policy/flow confirmation (Fed communication, realized liquidity easing, or explicit FX/monetary intervention) or for post-data reassessment around core PCE/GDP releases. | -4.0% | 2026-06-04 |
| ECB Eurozone Rate Cycle ecb_eurozone_rate_cycle | IYR iShares U.S. Real Estate ETF | Contested | 0.30 | Maintain a contested stance on IYR (no clear long/short), but with reduced conviction. The prior view relied in part on reported inflows (from 14 Aug) and the transmission chain: weaker macro → central bank dovish shift → benefit to rate-sensitive assets. That transmission requires confirming non-price signals (net inflows and softer macro). Price evidence only partially reflected the thesis (price_in excess = -2.55%; etf_window cumulative +1.24%; max drawdown -4.37%), so gains were limited. Crucially, independent non-price indicators have moved against the dovish-pivot story: since 2026-08-17 IYR shows net creation/redemption outflows of -$110M (-2.41% AUM) and 30‑day net flow = -0.94% AUM (contradicting the prior inflow assumption); Eurozone manufacturing PMIs surprised to the upside (EU manuf 52.8 vs exp 51.8, σ=+1.2; DE manuf 54.1 vs exp 52.1, σ=+2.0), reducing the likelihood of an imminent ECB dovish pivot; Fed‑related/system liquidity fell by ~$125B over 4 weeks, adding background pressure on risk assets. Option skew is relatively elevated versus SPY (+2.70pt) though IV percentile is low (0%), so option markets do not provide a clear long signal. Therefore keep the case contested but lower confidence from 0.40 to 0.30 and await clearer ECB signals or a material reversal in fund flows before enlarging exposure. | -2.5% | 2026-06-11 |
| Trump 2026 Tariff Policy trump_tariff_2026_policy | XLB Materials Select Sector SPDR Fund | Short | 0.30 | Maintain a medium-term (1–3 months) short bias on XLB. Arc window cumulative returns show XLB +1.20% / SPY +3.80% → XLB-SPY excess = −2.60%; the 9-day etf_price_window cumulative = +0.56% with max drawdown −2.74% (price_in is currently None and therefore not a standalone direction signal). The decisive non-price quantitative evidence continues to favor downside: US retail sales actual −0.6% vs exp +0.1% (σ≈−2.8, major cold); system liquidity (WALCL−TGA−RRP) fell −$191B over 4 weeks (as of 2026-08-12); XLB short interest rose +8.8% with days-to-cover = 1.37 (7-period cum +19.3%)—all direct signals of weaker demand/financing marginal conditions. Commodity-side support (CFTC copper net speculative long ≈ 75,758 contracts; steep WTI backwardation) provides background support for raw-material prices but does not by itself overturn the demand/liquidity transmission that hurts materials equities. Historical arc signals show split outcomes (roughly 12 long / 12 short in recent event sample), indicating high dispersion. Net: downside transmission remains the cleaner causal channel, so remain short but reduce conviction because the new trigger (#3992) introduces credible commodity-side counter-evidence and short-term price resilience. | +0.8% | 2026-06-20 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | EWY iShares MSCI South Korea ETF | Long | 0.30 | Maintain a 1–3 month long on EWY but with reduced conviction. The arc causal chain remains: a US macro slowdown would lower the odds of further aggressive Fed hikes (or delay tightening), favoring emerging-market/Korea risk assets and supporting EWY. Non-price quantitative support includes the prediction market P(hike)=0.47 (manifold) and a 30-day net subscription of +7.37% AUM (19-day window), indicating no strong market bias toward additional hikes and longer-window fund inflows into the ETF. Offsetting evidence has strengthened: price_in excess = -9.7% (EWY -12.57% vs EEM -2.87), arc-window EWY cumulative -16.63%, etf_price_window cumulative -12.57% with max drawdown -34.21%. Importantly, since 2026-08-17 there have been short-term net redemptions of -$266M (~-0.96% AUM) and Fed net liquidity WALCL−TGA−RRP is $5,792B (4w -$125B), a tangible liquidity headwind. The current trigger—China–Indonesia 2+2 ministerial meeting—is diplomatic/defense in nature (Xinhua) and shows no "on-point" non-price evidence (no FX intervention statements or reserves moves), so it is neutral to the EWY causal chain. Net: structural non-price signals still provide some basis for a long, but recent fund outflows and declining Fed liquidity materially weaken the case. Keep long, reduce conviction from 0.38 to 0.30, and wait for clearer non-price triggers (sustained net inflows, Fed liquidity relief, or a clear macro inflection) to increase exposure. | -13.7% | 2026-06-04 |
| Fed 2026 Rate Cycle fed_2026_rate_cycle | IWM iShares Russell 2000 ETF | Contested | 0.30 | Maintain contested. Prior assessment (2026-08-21) correctly flagged a split between dovish macro/repricing paths and extreme positioning/liquidity tail risk. Today's trigger (event_opp #4165 — a private chip startup raising ~$2bn and poaching Nvidia talent) does not resolve that split. Price-derived signals show IWM outperformance (price_in excess = +2.68%; arc etf_window cumulative +6.5%, max drawdown -3.95%, 8d cum +0.58%) and recent net inflows over 30 days (+0.65% AUM), which can be read as tape confirmation. But non-price, causal signals remain conflicted: CFTC COT (Russell_2000) net = -99,786 contracts (0th percentile, accelerating net-short over 4 reports), system liquidity WALCL−TGA−RRP down -$125bn (4w), and predictive market probability for a 2026 Fed hike = 0.47. The event itself is a private financing / talent move; per financing-event guardrail, such news is not a reliable directional driver for a broad small-cap ETF without matching non-price causal data (e.g., industry capex or supply changes). Therefore no structural evidence to flip or strengthen the prior view — remain contested with the same confidence. | +5.0% | 2026-06-03 |
| China Financial Regulation china_financial_regulation | EWT iShares MSCI Taiwan ETF | Long | 0.30 | Maintain a medium-term long on EWT. The primary quantitative supports are: 1) arc-window outperformance (2026-07-28→2026-08-18): EWT +11.11% vs EEM +4.78%, excess +6.33%; 2) persistent fund inflows: 30-day net inflow +5.33% of AUM (as of 2026-08-20). These non-price facts point to continued demand for the theme. Offsetting signals exist: the 6-day price_in excess = -1.22% (EWT -2.59% vs EEM -1.37%) shows near-term relative weakness; options and positioning are mixed — the current event shows much larger call OI than put OI (put_oi 16,788 vs call_oi 51,655, put_call_vol_ratio ~1.08) while prior windows recorded put_crowded readings, indicating inconsistent market positioning and reversal risk; short interest has risen (short_shares +37.6%, days_to_cover=1.19), increasing potential volatility and squeeze dynamics. The trigger (broker capital increases to alternative investment subsidiaries and deepen private equity/placement activity) is a financing/expansion event; per rule, financing alone is not an automatic bullish signal because effects depend on dilution/use of proceeds and balance-sheet implications. Net: non-price evidence is mixed but inflows and window outperformance remain dominant supports and no new structural contradiction emerged, so keep long at confidence=0.30 (no add). | +7.7% | 2026-07-29 |
| US-Iran 2026 War us_iran_2026_war | JETS U.S. Global Jets ETF | Contested | 0.30 | Maintain "contested." Two countervailing channels remain: (1) multiple first_signals pointing to a possible OPEC+ supply cut—if binding, higher jet fuel costs would be a clear negative for JETS; (2) prediction markets and macro risk indicators price a relatively benign near-term conflict scenario, which supports airline demand. Non-price quantitative evidence: polymarket yes_prob that a ceasefire continues through 8/31 = 0.94 (up from ~0.69 in the prior assessment), VIX/VIX3M ratio = 0.818 (calm contango), HY OAS = 2.75% (no acute credit stress). Price cross-checks show mixed signals: price_in_excess (48d) = -1.72% (JETS trailing SPY), while the arc window cumulative return (2026-06-11→2026-08-18) is JETS +4.66% vs SPY +4.02% (excess +0.64%); window max drawdown ≈ -12.7% indicates material downside risk. Flow and positioning: 30d flows +2.65% AUM (modest inflow), short interest +10.2% with days-to-cover = 3.23 (shorts have accumulated). The new conflict_escalation event is primarily narrative—no binding OPEC decision or supply/stock data—so it does not resolve the opposing channels. With no decisive non-price signal to pick a side, the call remains contested at the same confidence. | -1.7% | 2026-06-11 |
| US-Iran 2026 War us_iran_2026_war | EWG iShares MSCI Germany ETF | Contested | 0.30 | Remain contested. Today's reporting that the U.S. could impose a naval blockade on Iranian ports and hold it 'indefinitely' strengthens the theoretical transmission (conflict → shipping disruption / sanctions → supply-chain/commodity second-order effects → European manufacturing / German equities (EWG)). However, per system rules the direction must be supported by non-price quantitative evidence. Current non-price signals do not indicate the market has priced a sustained negative shock to German equities: prediction market (polymarket) shows 'US ceasefire through Aug 31' yes_prob=0.94 (high short-term ceasefire probability); EWG 30-day creation/redemption flows +4.57% AUM (capital not fleeing); reported short interest down −23.9% with days-to-cover=2.95 (no expanding short pressure); option ATM IV at 26th percentile (low fear). Price is only a cross-check: arc window cumulative EWG +3.73% vs EFA +4.90% → excess −1.17%; 53-day etf_price_window cum 4.96%, max drawdown −4.85%, price_in excess −0.88% (EWG 4.96% vs EFA 5.85%). Many recent event_opps are labeled first_signal, so historical-analog strength is weak. Conclusion: the new report matters conceptually but lacks the requisite non-price, arc-specific causal evidence to move to a directional call — keep contested. | +1.1% | 2026-06-05 |
| ECB Eurozone Rate Cycle ecb_eurozone_rate_cycle | XLU Utilities Select Sector SPDR Fund | Short | 0.30 | I acknowledge the prior assessment (2026-08-18, contested, confidence=0.20) correctly emphasized the causal chain — ECB hawkishness → higher real/nominal rates → duration-sensitive utilities underperform — and rightly remained cautious because of limited direct ECB-specific evidence. In this refresh I retain that structural chain but identify independent, non-price EU-side signals and fund flow data that strengthen a downside case. Quantitative evidence: 1) arc price_in_excess = -8.82% (XLU -5.25% vs SPY +3.57%), indicating notable cumulative underperformance; 2) XLU 1-month = -7.60% and 2026-08-21 1d = -2.28% (price moves are corroborative, not sole drivers); 3) eurozone manufacturing PMI prints are stronger-than-expected — EU manufacturing PMI 52.8 vs exp 51.8 (σ=+1.2) and Germany manufacturing PMI 54.1 vs exp 52.1 (σ=+2.0) — a direct EU non-price datapoint that supports higher ECB hawkish tilt and thus higher discount rates for long-duration utilities; 4) XLU AUM net outflows since 2026-08-13 = -$165M (-0.73% AUM), evidencing capital withdrawal; 5) option IV percentile is low (26%), indicating cheap protection and limited market fear. These non-price signals map cleanly to the arc’s causal chain (EU data → ECB hawkish expectations → duration pressure → XLU downside), so I move to an explicit short view while keeping conviction moderate because there is no single decisive ECB action and historical analogs show mixed magnitudes. | -8.8% | 2026-07-18 |
| Private Credit & Shadow Banking Stress private_credit_shadow_banking_stress | BIZD VanEck BDC Income ETF | Contested | 0.30 | Maintain prior 'contested' view. The previous thesis leaned on private-credit/BDC credit deterioration plus an outsized US retail-sales shock as the structural bearish channels, but that was partly offset by BIZD flows and price action. Since the prior update (2026-08-18) there has been no new independent non-price event that strengthens the bearish case. Price has moved: arc window cumulative BIZD +6.58% vs SPY +2.21% (excess +4.37%), indicating market has priced some relief. Non-price signals remain mixed: HY OAS is 2.75% (5d +4bp, 20d -2bp) and sector non-accruals stay elevated at 3.1% but show no clear worsening; fund flows since 2026-08-13 are small net outflows -$3M (-0.20% AUM) with 30d -0.05% AUM, and FINRA short interest fell to 2,770,338 (-6.9%) with days-to-cover=1—these flow/position metrics counterbalance a clean bearish interpretation. Therefore keep contested and do not raise confidence. | +4.4% | 2026-07-12 |