U.S. stocks ended October with a round of sharp fluctuations. Brian Garrett, a derivatives strategist at Goldman Sachs, identified seven "repair" signals in the market in his latest weekly report, suggesting that the current market positioning and sentiment have provided grounds for contrarian buying. However, he also admitted that the real test is not over yet: "We just need to get through 2026 first."

The Nasdaq 100 index fell by about 7% in October, marking it as the second worst-performing month since the start of AI's performance in 2023, and it underwent its third adjustment of the year. Chip stocks tumbled 7% to 8% on Friday, recording the third largest single-day decline in over 15 years; after remaining above the 200-day moving average for 355 consecutive trading days, the Philadelphia Semiconductor Index broke below that average on Friday and is currently testing its July lows. The S&P 500 index fell 0.6% for the week, closing at 7,580 points.

However, Garrett points out that despite the continuous pressure on spot prices, "some contrarian indicators are emerging, suggesting that the current trend may have certain corrective characteristics." Data from Goldman Sachs' commodities brokerage ( Prime Brokerage ) shows that hedge funds' current net exposure levels are historically associated with pullbacks of over 20%, while the S&P 500 has only corrected by about 3% from its historical high. This divergence in itself constitutes a contrarian signal.

The 355-day record for chip stocks comes to an end, with valuations returning to the low point of the AI cycle.

Goldman Sachs experts Peter Callahan pointed out that the Philadelphia Semiconductor Index closed below its 200-day moving average on Friday, ending a streak of 355 consecutive trading days above this level, which was the third longest consecutive strong period in the past 10 years. The index is now pulling back towards its July lows. The forward price-earnings ratio for the semiconductor sector has fallen to around 16 times, reaching the valuation low point of this current AI cycle.

What triggered this round of sharp declines was a week of intensive financial reporting season. Callahan believes that the relevant data points "to some extent show signs of a turning point in growth rates," but he emphasizes that these signals "all come from outside of the AI core complex." He also cites Amazon's forecast for its revenue visibility over the next two years as evidence to support this view.

The following day, the reaction was even more intense: NOW fell by 25% in a single day, INTC fell by 30%, RDDT fell by 30%, and FFIV fell by 20%. Many individual stocks recorded their worst performance since their financial reports were released. Callahan attributed the reasons to overcrowded positions, a compressed financial report calendar, and ongoing structural industry disputes.

Garrett further pointed out that the direction of market fluctuations has shifted – stocks are 'no longer soaring upwards.' Taking Intel as an example, in June of this year, its average daily increase reached 6.5%, while the current average daily decline has already exceeded the average increase. Semiconductors ETF and QQQ both show the same pattern: the average amplitude on down days is once again greater than that on up days. In Garrett's words, stocks are 'operating again like stocks – downward fluctuations are more extreme than upward fluctuations.'

Index volatility makes up for the losses, but Gamma remains unchanged.

The first reversal signal listed by Garrett is that the implied volatility of stocks at the macro level has finally begun to "catch up" with the volatility of individual stocks and the credit market, which had already fully priced in panic sentiment prior to this.

A weekly report from Goldman Sachs' derivatives team indicates that the implied volatility of the Nasdaq 100 index surged by nearly 2 volatility units within a single trading day. For the first time, clients sought to buy protection against index declines, and there was a change in the flow of put options on the semiconductor sector. The S&P 500 index briefly fell below the medium-term CTA (trend-following strategy) threshold during trading. The deviation between the actual volatility and the implied volatility of QQQ was one of the most significant single events in the past year.

The volatility of volatility also soared simultaneously: VVIX reached a high of 124 during trading, the highest level since the outbreak of a specific geopolitical conflict, before subsequently falling back. On Friday, the daily trading volume of U.S. stock options approached 100 million contracts, setting a new record for the fourth-highest volume in history.

The only one who remained inactive was Gamma. Garrett pointed out, "Despite the significant actual daily fluctuations in the index, there was no obvious reaction in the pricing of the 0DTE straddle options." Taking the S&P 500 as an example, the cost of the straddle options, which were influenced by both Apple's financial report and the mid-term elections, was only 150 basis points—lower than the actual decline of 180 basis points on Friday alone. In terms of the implied volatility for the year-end direction, the S&P 500 had ±5.1%, the Nasdaq 100 had ±7.8%, and the Russell 2000 had ±6.3%, indicating that the volatility of the larger market was once again higher than that of the smaller markets.

Hedge fund net exposure reaches ‘bear market level’, but indices do not fall

The second contrarian signal comes from Goldman Sachs Prime Brokerage data: hedge funds made net purchases of U.S. stocks for the first time in four weeks (+1.1 standard deviations), but 73% of these purchases were in macro products, mainly consisting of short covering. The number of open short positions in U.S.-listed ETF securities decreased by 4.8%, representing the largest single-week covering volume in over six months, with technology, large-cap equities, and credit ETF leading the way.

Although technology was the sector with the largest net purchases that week, according to Goldman Sachs' block trading team, this portion of buying (as of Thursday) only accounted for about 40% of the selling volume over the past three weeks. The "Big Seven" (Mag) have experienced net selling on four out of the past five trading days, "indicating that risk appetite has not substantially improved."

It is worth noting that Garrett pointed out that Mag 7 remains one of the "fields where minority positions continue to accumulate" within the current TMT. The long positions are still close to their highs for the year – while the market is selling off technology stocks, it seems to be reluctant to part with those sectors where positions are most concentrated.

More crucial data is that the net leverage ratio of U.S. fundamentals has dropped to 47%, which is in the 7th percentile of the past year. Garrett conducted a regression analysis of this data with the S&P 500 drawdowns and found that this level of net exposure usually corresponds to drawdowns of over 20% in history – yet the current index has only fallen by about 3% from its historical high. In other words, hedge fund positions are already preparing in advance for a “bear market that has not yet occurred”, which precisely constitutes a basis for a contrarian bullish view.

80% of constituent stocks rose, yet the index closed lower: "Defensive stocks are too expensive"

The fourth and fifth signals come from the internal structure of the market. On Thursday, 406 of the constituents in the S&P 500 index closed higher, which is one of the highest proportions on a single day since 2026, yet the index still closed lower that day. Garrett created a scatter plot comparing all days of decline with the proportion of stocks that rose on those days, and the data point for Thursday stood out alone—80% of the constituents were rising while the index was falling, which is almost an unprecedented case in history.

Behind this is the ongoing divergence between AI and non-AI stocks: Goldman Sachs' AI winner portfolio continues to face pressure, while the S&P 500 (S&P ex-AI), which excludes the AI sector, is expected to record its best single-month performance since 2022 this month. Meanwhile, U.S. Treasury bonds ( TLT ) and the AI winner portfolio exhibit a two-week correlation of around -0.7, near the lowest levels of the past two years, reflecting the classic 'stock market decline, bond market rise' safe-haven pattern.

In terms of valuation, the S&P 500 has seen a cumulative increase of 10.7% over the past year, but its forward price-earnings ratio has fallen back to its lowest level since 2026, at around 19 times. This situation, where the index rises while valuations “catch up” with earnings, constitutes a contrarian signal. However, “defensive stocks are already quite expensive”: the dividend yield factor is at the 90th percentile of the past decade, the valuation factor is at the 80th percentile; whereas the momentum factor is only at the 30th percentile, and the growth factor is at the 40th percentile. The forward valuation multiples of low-volatility indices have expanded to the level of growth stocks.

Callahan writes that conversations about "diversifying portfolios" are on the rise, but "easier said than done." On trading days not led by AI sectors, there is often a sense of unease as positions seem to be dismantled, and it's difficult to find non-AI targets that are not disturbed by the prevailing narrative of AI. He listed some currently valid individual stocks, including: AAPL, SNPS, ADI / TXN / NXPI, SPOT, V/ MA, EBAY, and LYV.

Apple sets a new historical valuation high, with discounted pricing for the mid-term elections

This week, there is a high concentration of market catalysts: Apple's ( AAPL ) post-market financial report on Monday, the financial reports of AMD and ANET on Tuesday, the mid-term elections on Tuesday ( Kalshi with a predicted 65% chance of a Democratic Party landslide victory), details of the Treasury Department's refinancing on Wednesday ( QRA ), and Friday's non-farm payroll data (29,000 jobs added last month). The implied volatility for the S&P 500 this week is 1.84%.

Callahan Focus on five individual stocks. Apple has reached a new historical absolute valuation with a forward P/E ratio of 36 times, which is 17 times higher than the S&P 500, setting a record for relative valuation. This super-large-cap stock, which is still considered a safe-haven asset, is also the most expensive ever, confirming the logic that "defensiveness comes at a cost."

Regarding the mid-term elections, Garrett expects them "not to be a major event" for the overall index, but sees it as an opportunity for the thematic basket strategy to shine. He specifically mentioned the pair trades that deserve the most attention: going long on renewable energy and grid-related sectors, while also buying into the downside protection of ultra-large-scale data centers and data center equipment sectors – "Who will pay for the electricity used by data centers?" He considers this to be "the most undervalued election-sensitive theme" in the U.S. stock market.

Last two signals: The financing expectations for AI have reversed, and retail investors are now adopting a strategy of "buying only after asking questions".

The last two contrarian signals of Garrett are relatively flexible. Sixthly: experts in the credit market have begun to suggest that the financing needs of AI may be lower than expected. If large cloud providers reduce their dependence on the bond market, it could help alleviate market anxiety, but "it will take time for this to materialize."

Seventhly, and this is what Garrett talks about with great enthusiasm: the behavior of his university classmates and father-in-law has begun to send out warning signals – this group has shifted from buying on dips without hesitation to asking a question first before making a purchase. Garrett believes that retail investors are on the eve of an “agent-based transformation,” and he cites an advertisement as an example: during the break when children go out to ask for candy, the advertisement promises to “set up your intelligent agent options strategy in just 7 minutes.”

Overall, Garrett believes that the signals at the positioning level support a correction: the net open positions have corresponded to pullbacks of over 20% in history, there are panic-induced abnormalities in market breadth, defensive stocks are severely overvalued, and retail investor sentiment is wavering. However, the flow of funds indicates that deleveraging is far from over: trading desks are still selling tech stocks, Mag's long positions are still near their annual highs, and CTA's system is about to trigger a new round of short-selling signals. Moreover, the options market's pricing for the combination of Apple's financial report and mid-term elections is even lower than the actual volatility costs on Friday alone. As for 2027, the options market has priced the internal correlation of the index at a historical low of around 20%—by then, the index may remain stable, while the internal situation could be undergoing a complete upheaval.