Dealer Gamma Has Fallen Off A Cliff: Chief Goldman Equity Trader Warns 'Brace' For More Pain Into Month-End

On February 28th the U.S. and Israel launched a surprise attack on Iran marking the start of the war.

Most U.S. investors’ initial reaction was something along the lines of "this will likely be another temporary disruption... a 'Venezuela 2.0'".

This initial read has turned out to be a clear miscalculation.  

We are 20 trading sessions into March with only 2 to go, and as Goldman Sachs head of equity execution, John Flood, writes in his latest 'Mark to Market' note, Q1 has been one of the hardest / longest that I can remember.

The S&P 500 has lost 7.41% MTD (down 696bps on the year).

Surging oil prices, interest rates, and geopolitical uncertainty have dragged the S&P 500 P/E multiple from 21x a month ago to 19x today.

US 10Yr yields were at 3.9% on 2/27 and have since spiked to 4.43%.

Our work shows if 10yr yields move 2SD higher (~50bps) within a one month period it becomes very difficult for the stock market to absorb...we are officially there. 

Market trading volumes have exploded since the start of the war with a month to date daily average of 20.2b shares trading across all US equity exchanges. For context the daily average January – February was 19b shares and 2025’s daily average was 17.5b shares. Despite an explosion in trading volumes liquidity has been hard to come by. S&P 500 Futures top of book depth (how much notional a trader can execute at price that is shown out loud on the screens) has lived below $7mm since the start of the war vs longer term historical average of $15mm. A clear signal that extreme stress still exists in the system. 

There are now 1000 more listed ETFs (~5k) in the US than listed stocks (~4k). Institutional investors have been using ETFs as hedging vehicles – mostly on the short side to dampen headline risk. ETFs typically represent ~30% of the total US equity volumes. However, over the the last 3 weeks ETFs account have accounted for ~40% of the total tape. This is another signal of stress in the system and a warning to brace for continued volatility at the index level. 

Until we see top of book depth trend north of $10mm and ETFs represent <33% of total market trading volumes, we will remain in a very fragile tape. These two are related because as top of book liquidity in futures reduces, ETFs are leaned on even further as a way to both add and reduce macro risk. However, when transacting in block form, lower top of book liquidity (coupled with higher vol) can lead to wider risk markets across the street -- an important dynamic to keep on the radar if long only trading activity picks up. 

Source: Bloomberg, Goldman Sachs Global FICC & Equities as of March 24, 2026. Past performance is not indicative of future results. H/T Chris Lucas

Since the start of the war long only trading activity (specifically asset managers and SWFs) on our desk has essentially been nonexistent (aside from some one-off situations). It is eerie. The recurring word being thrown around is frozen. I worry that we are now approaching the point in this conflict where the L/O community will become unfrozen and start cutting some real risk. The first place I am watching for noteworthy sales is European stocks that US L/Os bought last year and have been sitting on ever since. LIFO.   

Thankfully corporates are buying the weakness. There has been lot of debate around hyperscaler capex explosion and how it will impact buybacks. Corporates will eventually have less cash in their pockets to buyback their own stocks but we haven’t seen them turn this repurchase dial back yet. The last 3 weeks have been one of the most active periods in the history of our buyback desk at GS (from a notional executed perspective). If we start to see these repurchases dwindle it will be a cause for further concern. 

US Pensions are now modeled to BUY ~$20bn of US equities for quarter-end...hopefully this can absorb some of the bleeding. The systematic community is also running out of steam on the supply side of the equation. Through Thursday’s close, we estimate that this cohort has sold -$85bn of US equities over the last 30 sessions.

We now have CTAs short -$37bn here. Asymmetry lives to the upside -- over the next month, we estimate CTAs are buyers in every scenario.

Clearing month-end should be significant from a technical lens.

Dealer gamma has fallen off a cliff and is now at peak-short levels.

As of Fridays close, dealers are short over -$7bn of gamma (the second lowest reading on record, via squeezemetrics).

We expect this short gamma pocket to roll off at the end of the month. Until then brace for more procyclical hedging AKA buying on up moves and selling on down moves. 

Source = GS GBM via squeezemetrics H/T Cullen Moran

Finally, in terms of the HF community some signs of capitulation are starting to emerge in our prime data. 

Last week HFs net sold US equities for a 6th straight week and at the fastest pace since Apr ‘25 (-1.6 SDs 1-year), driven by short-and-long sales in Single Stocks and to a lesser extent short sales in Macro Products.

Source = GS PB H/T Vincent Lin 

On a trailing 6-week basis, the recent US net selling by hedge funds is the 3rd largest over the past decade and starting to approach the levels seen in Apr-May ’20 during Covid and (to a lesser extent) into Liberation Day.

Source = GS PB H/T Vincent Lin 

US Fundamental L/S Net leverage fell -3.1 pts last week – the sharpest weekly reduction since early Apr ’25 (week of Liberation Day).

Source = GS PB H/T Vincent Lin 

Korea was among the most net bought markets by hedge funds globally in January/February but is among the most net sold so far in March.  Approximately 70% of the YTD $ net buying in Korea coming into March has now been reversed on the Prime book.

Source = GS PB H/T Vincent Lin 

From a technical perspective, investor positioning last week took a large step toward risk reduction (driven by the HF community).

The last several weeks have been characterized primarily by hedging activity, with little selling of core portfolios.

Last week, however, our US Equity Sentiment Indicator dropped this week to -0.9, reflecting a large outright reduction in equity exposure.

Sentiment Indicator levels below -1 have historically been predictive of above-average equity returns, although the signal improves when the indicator drops below -1.5.

Closing out from a fundamental perspective, S&P 500 earnings should continue to grow at a solid rate this year, barring a severely prolonged disruption.

Our base case is for 12% S&P 500 EPS growth in 2026.

While EPS growth would be modestly weaker in the adverse scenarios outlined by our commodity strategists and economists, AI investment represents a key tailwind, accounting for 40% of S&P 500 EPS growth this year.

During the 12 months following past oil supply shocks, S&P 500 EPS growth has ranged from -15% following the Iraqi invasion of Kuwait in 1990 to +18% during the Second Gulf War in 2003.

Professional subscribers can read much more from Goldman's Sales & Trading team here at our new Marketdesk.ai portal