Goldman Sachs Trading Desk: Five Signals Flashing, US Stocks Enter “Post-New-High Pullback Window”

```

After experiencing severe volatility in April, the S&P 500 has quietly returned to historic highs, but an alarm has been sounded within Goldman Sachs' trading desk.

Goldman Sachs trader John Flood explicitly pointed out in his weekend report that although he believes the S&P 500 will be significantly higher than current levels for 2026 as a whole, five major warning signals are flashing simultaneously, and the trading desk is preparing for a near-term pullback in the index.

At the same time, another Goldman trader, Brian Garrett, noted unusual moves from hedge funds deleveraging on a large scale in his “Weekend Memo” report. Both believe that if a pullback occurs, it should be viewed as a buying opportunity.

The timing of this warning is quite subtle—this will be the busiest week of the year, with both the Federal Reserve and Bank of Japan announcing rate decisions. Around 44% of the S&P 500’s market cap will report earnings this week, including Google, Microsoft, Amazon, and Meta after Wednesday, and Apple after Thursday.

Signal One: Hedge Funds Hit the Brakes, Largest Weekly Deleveraging in Seven Months

Goldman’s prime brokerage data shows that overall trading activity fell last week for the first time in 13 weeks, with U.S. equity nominal deleveraging the largest since seven months ago (September 2025), mainly driven by risk unwinds. Sector-wise, the consumer discretionary and tech sectors saw the most aggressive deleveraging—the third largest weekly drop in five years.

Flood points out this is the first of five warning signs: After large-scale covering of macro short hedges, gross leverage has significantly contracted. U.S.-listed ETF shorts fell another 1.4% last week and are down 21.5% for the month, leading the way in credit, information technology, and small-cap ETF short covering.

Garrett described hedge funds’ net exposure as “remaining relatively restrained within a ±53% range for the year,” characterizing it as “prudently managing risk in a market environment of frequent ‘unknown unknowns’.”

Signal Two: End-of-Month Pension Rebalancing Pressure Hits Record

The second signal comes from passive selling. Goldman estimates that April end-of-month pension rebalancing will generate about $25 billion of U.S. equity selling, ranking among the top 15 since 2000. Excluding quarterly factors (including both monthly and quarterly rebalancing), this would be the largest single-month estimate ever.

Measured in absolute terms, the $25 billion of selling ranks at the 83rd percentile of all buy/sell estimates over the past three years, and at the 92nd percentile since January 2000. Comparable non-quarterly selling estimates in history are: about $20 billion each in November and April 2020, about $19 billion in May 2025, and about $18 billion in October 2022.

Signal Three: CTA Buying Exhausted, Turning to Potential Sellers

The third signal points to trend-following strategies (CTAs). Goldman’s futures desk shows that since April, CTAs have been the most important driver of global equity gains, buying about $53 billion in global stocks and net buying about $32 billion in S&P 500 alone this month (with $23 billion bought last week).

However, this buying momentum has ended. Flood notes this is the first time in over a month that CTAs are no longer net buyers in the S&P 500—in stable conditions, a slight bias toward selling, and in a downturn, more significant supply pressure. This means the largest marginal buyer previously supporting the market is now “fully loaded and waiting,” and the market has lost an important automatic stabilizer.

Signal Four: Market Breadth Severely Worsens, Hidden Risks Behind New Highs

The fourth signal reveals structural risks beneath the surface prosperity. Last Friday, the S&P 500 closed at a fresh all-time high, but market breadth that day was the second worst ever for a new high: 324 constituents fell, with net breadth at -148. The only worse record was in October 2025, when 80% of the constituents fell on a new high day.

The severe divergence in market breadth means the rally is highly concentrated in a few heavyweights, and overall participation has shrunk markedly. Historically, such structures are often precursors to high-level volatility or corrections.

Signal Five: Sentiment Indicator Enters “Stretched Zone,” Semiconductor Rally Becomes a Concern

The fifth signal comes from Goldman’s U.S. equity sentiment indicator: Investor positioning shows a “stretched” characteristic. Overall sentiment framework still reflects relatively elevated positions.

Meanwhile, Flood specifically flags the extreme moves in semiconductors. The Philadelphia Semiconductor Index (SOX) has risen for 18 consecutive trading days, marking the longest streak in history, with Friday’s close about 50% above the 200-day moving average—the most extreme deviation since the bubble peak in 2000 (then SOX was over 100% above the 200-day). This trend is driven by increased confidence in AI computing power and monetization, and fresh worries about AI supply constraints.

From the derivatives market, the S&P 500’s gamma positioning is at a rare level, with market makers strongly net short gamma on spot breakouts, meaning directional moves will significantly amplify volatility. Few professional investors currently hold outright long positions, July call option implied volatility is trading near 12, and bullish trades remain a “lonely trade.”

Despite the five warning signals pointing to a short-term correction, Goldman still believes the S&P 500 will end 2026 well above current levels, and any pullback should be viewed as a structural buying opportunity. Historically, since the financial crisis, whenever the S&P 500 retraced more than 10% then regained previous highs, subsequent 1-week, 1-month, and 3-month average returns were 1.5%, 5.2%, and 8.6% respectively.

Risk Warning and DisclaimerMarkets are risky, investments must be cautious. This article does not constitute personal investment advice and does not take into account the individual investment goals, financial situation or needs of any particular user. Users should consider whether any opinions, views, or conclusions in this article are appropriate for their particular circumstances. Investing based on this information is at your own risk. ```