The Most Worthwhile Review of This Week

What stands out this week is not any single day’s wild swings, but rather the clearer definition of three main themes: increased pressure on the broader market, a more concentrated focus on semiconductors, and renewed emphasis on gold and long-term bonds due to banking pressures. ETF Radar’s most notable changes this week align with these three lines.

Key Changes This Week

• The broader market has acknowledged weakness: SPY has fallen about 2.86% over the past 20 days, and QQQ has dropped around 3.14% in the same period; this month's tape suggests a generally weak performance for US equities. - SPY remains negative but with a higher concentration of evidence, moving from -0.46 to -0.68. - GLD has shifted from neutral to positive, increasing from 0.34 to 0.64; this is one of the most significant directional reversals seen this week. - TLT has moved from negative to mixed, indicating that long-term bonds are no longer unilaterally bearish, with bank stress and high-interest rate bets now pulling in opposite directions. - IWM has also shifted from negative to mixed, suggesting a weakening of the original interest-rate pressure on small-cap stocks. - MCHI and FXI have both moved from positive to negative, indicating a clear cooling off for Chinese large-cap indices this week. - The semiconductor theme continues to strengthen: SOXX, SOXQ, CHPS, and DTCR showed notably stronger signals this week. - A new thematic opportunity entered the operational state this week, with GLD being the most important; opportunities related to TLT are also now in an operable state but remain mixed rather than one-sided. - Conversely, evidence for COPX has become weaker, although its directional label remains positive, it is approaching a status where it should not be treated as a primary theme.

This Week's Main Story 1

Chips are still shaken, but the main trend is firmer.

The markets this week appear quite alarming on the surface. Semiconductor ETFs like SOXX, CHPS, and SMH saw significant single-day declines. However, if we look at the past 20 days, SOXX is up about 4.35%, CHPS is up about 6.05%, and SMH is up about 2.71%. In plain terms, while short-term volatility is evident, the long-term trends remain intact.

What happened this week? The core was several industry news stories corroborating each other. On one side, SK Hynix raised a substantial amount of capital, approximately $29.65B. Raising capital means the company is putting its 'funding and production plans' on the table. On the other side, discussions about memory crunch, or 'memory supply shortage,' continued to heat up. In plain terms, this means that while expanding capacity isn't happening as quickly, AI-related demand is still pushing forward. Additionally, messages from TSMC and Foxconn along the supply chain indicate that there has been no sudden slowdown in demand for AI servers, cloud hardware, and advanced chips.

Why does this affect the ETF? Because semiconductors are not just about 'demand', but also about 'whether supply can keep up'. If demand is strong and supply is slow, stronger players in the industry are more likely to secure better orders and profits. For example, if a cinema suddenly fills up but cannot add more screenings immediately, ticket prices and seat scarcity will rise. Therefore, the reaction of the chip ETF is not just 'AI is hot', but rather 'capital expenditures continue while supply fails to keep pace, making it easier for industry sentiment to persist.'

First, let's look at SOXX. The ETF Radar model is slightly positive towards SOXX with consistent signals this week. Its confidence level rose from 0.56 to 0.78. The primary reason for this increase is not the price movement itself but rather more comprehensive non-price evidence: large fundraising, production bottlenecks, and ongoing AI demand. Although it fell by 5.87% in one day, over the past month, the tape has been positive, indicating short-term correction with ongoing medium-term validation. Such retracements seem more like high-level oscillations rather than a refutation of the main trend.

Check CHPS. The reading is positive for CHPS with multiple indicators moving in the same direction. This week, it jumped from 0.38 to 0.70, a clear strengthening. CHPS can be seen as a basket more focused on the semiconductor chain. It has risen about 6.05% over the past 20 days, aligning with the system’s judgment. The signal strength increased this week mainly due to the supply tightness narrative shifting from 'market imagination' to 'companies are indeed raising funds and industries are replenishing capacity, albeit slowly'.

Thirdly, look at SMH. The model is slightly positive towards SMH with more focused evidence. It is actionable this week and thus in a state of being 'worth keeping an eye on'. Over the past 20 days, it has risen by approximately 2.71%, which positively corroborates our one to three month outlook. However, we must also acknowledge that there are signs of call IV at around 57% indicating some heat in the options market. IV stands for implied volatility, essentially an 'estimated thermometer' for future price movements according to the option market. High call IV suggests a demand for upward volatility but could also indicate crowded trading. Therefore, while this line is not without risk, it now has stronger industrial evidence and greater volatility.

Add a note for EWT. The data leans positive for EWT, with signals ranging from neutral to generally consistent. It is a Taiwan market ETF. This week's new verdict gave it a positive lean, due to direct support: $3.2 billion in financial guarantees from Google and approximately $1.6 billion in net inflows supporting the Taiwanese AI supply chain. However, over the past 20 days, EWT has only slightly approached equilibrium, so here it seems more like 'fundamentals are leading while prices have yet to fully catch up.'

The key point is not 'chips rising every day'. The focus is that more evidence this week indicates that capital expenditures for AI have not stopped, and the supply bottlenecks have not eased. Therefore, even if the broader market weakens, semiconductors remain the strongest sector currently. However, the stronger the trend, the greater the short-term fluctuations typically are, and this was evident this week.

This Week's Main Story 2

High interest rates are pressuring the broader market but supporting banks.

The second main trend is the divergence between the broader market and banks. This week, the negative signal for SPY has become stronger, but the bank-related ETFs have maintained a positive bias. This explains why many people feel that 'the index is uncomfortable, but some financial stocks are still doing well.'

What happened this week? The core issue is whether the Fed's path will be 'higher for longer'. In simpler terms, this means rates may stay at higher levels for a longer period. Supporting this view are the latest dot plot and SEP (Summary of Economic Projections). The SEP is the Fed’s projection of economic and inflation outlooks. Additionally, May's CPI was around 4.2% year-over-year, indicating that inflation has not retreated cleanly. Markets had hoped for a quicker easing, but this week's evidence did not provide such an easy answer. Consequently, there is increased pressure on the broader market, especially the more overvalued segments.

Why does this affect ETFs? Because interest rates are like the

First, let's look at SPY. ETF Radar is slightly negative on SPY with signals being consistent. This week, confidence rose from 0.46 to 0.68. Moreover, over the past 20 days, SPY has fallen by approximately 2.86%, and the tape in the last month is supporting this judgment. This does not mean that the market will continue to decline in a straight line, but at least the current evidence suggests that pressure from interest rates has not disappeared.

Now look at KRE. The ETF Radar model is slightly positive for KRE, with multiple pieces of evidence pointing in the same direction. KRE is an ETF for US regional banks. It remains actionable this week, having risen about 7.51% over the past 20 days. Moreover, within the system's window, it has also performed notably better relative to SPY. The underlying message is straightforward: while indices were initially hurt by high interest rates, banks can benefit from the spread dynamics. Of course, there are counter-arguments this week, such as 'reiterating anti-inflation does not necessarily mean more hawkishness.' Hawkish means tighter policy and less urgency to cut rates. These counter-arguments exist, so we cannot view banks as unidirectionally favorable. However, for now, non-price evidence still leans in favor of the bank sector.

Examine KBE and XLF. The signals are slightly more positive for KBE, with evidence that is more focused; they are positively inclined towards XLF but somewhat weaker than KBE. Over the past 20 days, KBE has risen by approximately 6.77%, while XLF has gained about 4.18%. Both align with the system's main trend. KBE appears to be a basket more heavily weighted towards banks. XLF encompasses a broader financial sector, including entities beyond just banks. Therefore, if 'higher rates for longer' continues to gain traction, banks would typically benefit more directly than the broader financial sector; however, if the market starts to worry about credit risk, the pure bank ETFs could experience greater volatility.

The key takeaway is that this is not simply about 'financials strong' or 'techs weak'. Rather, high interest rates are weighing on the broader market while supporting certain banks. This is the most typical sector divergence of the week.

This Week's Main Story 3

Banking pressure pulled gold and long bonds back again

The third line can be seen as a return to risk hedging. This week, GLD shifted from neutral to slightly positive, and TLT moved from negative to mixed. It is not due to a sudden new macro story, but rather because the old main thread of 'banking and credit stress' has gained new evidence.

What happened this week? The most critical data point is the market's prediction that the probability of three or more U.S. banks failing by 2026 (yes_prob) is approximately 0.67. Additionally, the gold COT net long position increased to 113,721 contracts. COT refers to Commitment of Traders reports, which essentially show where large funds are positioned in futures markets. In the long bond market, there is also a crowded short position on the 10-year Treasury futures. A crowded short position means that many have bet on the same direction. If there is a risk-off move or policy support, covering these shorts could happen quickly.

Why does this affect ETFs? If the market is concerned about banking system stress, there are two common transmission channels. One is a shift of risk-averse funds into gold. The other is a move to long-term bonds, or betting on regulatory and liquidity support that would lower long-term interest rates. Long-term interest rates are the rates for borrowing over extended periods. Lower rates typically lead to higher prices for long-term bonds.

First, let's look at GLD. ETF Radar is slightly positive towards GLD with a relatively strong signal. It has flipped from neutral to slightly positive this week, which is quite clear. But to be honest: GLD has fallen about 8.53% over the past 20 days and has been diverging on the tape for the last month. In other words, the system has turned bullish, but the market has actually been voting against it for the past month. This doesn't mean the judgment is invalid, but at least it suggests that this logic now more closely resembles 'non-price evidence first, with price not fully in agreement yet.' What we need to watch next is whether the bank stress line continues to escalate.

Now let's look at TLT. The model gives TLT a mixed assessment for now, with mixed evidence. On one hand, bank stress, crowded short positions, and safe-haven logic support TLT. On the other hand, market concerns about higher rates lasting longer put pressure on long-term bonds. So it is not yet a clear-cut single-sided conclusion. However, from the tape, TLT has risen by approximately 2.38% over the past 20 days, and in the last month, it has confirmed the 'safe-haven and unwind' logic. In plain terms, this week long bonds are no longer just 'not working because rates are high', but rather a tug-of-war between two forces.

When viewed together with the previous two, this narrative becomes complete: on one side, the broader market is under pressure; on another, semiconductors have a sectoral (mainline); and on yet another, bank pressures are giving gold and long-term bonds renewed relevance. Therefore, this week’s market is not single-threaded but clearly entering a phase of deeper (divergence).

Next week will be closely watched.

• Keep an eye on whether the three main lines of the ETF Radar continue next week: news about semiconductor supply shortages, and if there are any new company capital actions following up. - Continue to monitor whether there is new hard data confirming the Fed's path, especially in terms of the dot plot, SEP, and inflation-related interpretations, which should still support 'higher for longer'. - Keep an eye on new evidence regarding bank and credit stress, focusing on predictive market probabilities, COT positions, and any news related to liquidity support.

Conclusion

To sum up this week: the major indices have been more influenced by high interest rates, while semiconductors continue to be supported by industry evidence. Gold and long bonds have returned to center stage due to banking pressures. ETF Radar will next closely watch which of these three trends will secure confirmation first. Again, the above is a directional prediction from a system-based quantitative model for general markets and does not constitute personalized investment advice; please consult a licensed investment advisor before investing.