Hedge Funds Pile Into Short Bets as Goldman Flags Squeeze Risk

Hedge funds have added to their short positions in stocks until bearish bets sit near a five-year high, and Goldman Sachs has warned that a rebound could force a sharp squeeze, Bloomberg reported on October 22, 2026. A short position is a bet that a stock will fall. The fund borrows shares, sells them, and aims to buy them back later at a lower price.

The scale of the positioning is visible in Goldman's prime brokerage data, which tracks the lending and trading services the bank provides to hedge funds. Short leverage in the bank's prime book stood at the 99th percentile of its five-year history, according to reporting on the same week's positioning. Short leverage is the gross value of borrowed-and-sold stock relative to a fund's capital. A reading at the 99th percentile means the books are about as heavily short, measured against their own history, as they have been in five years.

A short only pays if the price falls, and the borrowed shares must eventually be bought back, or covered. When prices rise instead, that covering becomes buying pressure, which is the mechanism behind the squeeze risk Goldman flagged.

The shift has been building for months. In February 2026, hedge funds' short sales of U.S. stocks outpaced their long purchases two to one, per Bloomberg's February reporting.

Goldman's own assessment has moved in a different direction on one measure. In an October 9 note, Goldman stated that hedge fund positioning in equities for long-short strategies is at historically low levels, pointing to volatility in interest rates and oil.

Those readings can coexist. Net exposure, the gap between long and short bets, can be low while gross short exposure is high. A fund can be cautious on the market as a whole while hedging a concentrated set of longs with a growing pile of shorts. In my view, which of those two pictures dominates the tape right now is the question the squeeze debate turns on.

Tech is where the selling has been most pronounced. Goldman's prime desk said hedge funds retreated from technology stocks at the fastest pace in at least a decade.

Performance reflects the same stress. Global fundamental long-short funds dropped 3.8% in October 2026 through the latest Goldman figures, yet remained up 8.8% for the year as of October 23, according to Goldman Sachs as reported by Reuters. The month-to-date loss is still unfolding.

The year-to-date gain is the cushion that makes the current drawdown look more like a rotation than a rout.

Asia is a harder story. Asia-focused fundamental long-short funds fell 18.6% on average through July 28, according to a Goldman prime brokerage note reported by Reuters on July 30. The figure suggests stock pickers have had a difficult summer outside the United States.

The squeeze risk has a recent precedent. In January 2021, Melvin Capital Management lost 53% for the month after bearing losses from soaring prices of heavily shorted stocks, including a GameStop short, as MarketWatch reported at the time.

That episode was concentrated in a small number of names. The current warning is about crowding across the whole book, which is a different risk. A broad, index-level rally driven by covering would move prices for savers in index funds and retirement accounts, not just for the funds on the other side of the trade.

Morgan Stanley's January 2026 outlook stated that hedge funds, particularly market-neutral strategies, are attractive. Market-neutral funds carry matched long and short books, so they are designed to be less exposed to the direction of the overall market.

Heavy short books inside those strategies would look different from heavy net-short books in directional funds. That distinction matters for how any squeeze would transmit through the market.

The useful test for the next few weeks is whether short leverage falls through covering without a price spike, or whether a rally forces a scramble. Goldman's warning is a risk flag, not a forecast of a rally, and the positioning data describes what funds hold, not where prices will go. Anyone reading these numbers should keep that line clear.