Quant Long-Short Funds Down 3.6% in Two Weeks as Momentum Collapse Meets Korean Stock Turmoil

Miles Bennett
Published todayAbout 11 min read

Systematic long-short funds fell 3.6% in two weeks, surrendering roughly a quarter of their year-to-date gains in the worst drawdown since August 2025; Goldman Sachs points to a momentum unwind compounded by Korea's leveraged-ETF feedback loop — a stress test for every portfolio carrying momentum exposure.

01

How bad is a 3.6% drop in two weeks?

According to Goldman Sachs' internal trading-desk report, systematic long-short funds — hedge funds that use quantitative models to go long and short simultaneously — fell a cumulative 3.6% from June 22 to July 8. Year-to-date returns dropped from 14.4% to 10.8%, wiping out roughly a quarter of the year's gains.
This means → profits that took most of the year to build evaporated in two weeks, far faster than a normal pullback.
Warning signs appeared a week earlier: the five trading days ending June 29 saw a 3.1% loss, the worst five-day stretch since December 2023, with June 29 alone accounting for 1.0%.
02

Why did quant funds get hit all at once?

Goldman trader Mario Laicni attributed the losses to two forces: violent sector rotation inside the market + a collapse in momentum trades. In plain terms = the stocks that had been rising hardest suddenly reversed, and quant funds were the biggest holders of "chase the winners" strategies.
Losses concentrated on the short side — the stocks funds had bet against actually rose. U.S. equities were the biggest source of pain, followed by developed-market Asia and Europe.
This reflects a core vulnerability: when market style flips, the quant playbook of "long winners, short losers" gets squeezed from both sides — longs fall, shorts rise, and losses compound.
03

Why was Korea the epicenter?

Goldman flagged that single-stock leveraged ETFs on Samsung Electronics and SK Hynix have ballooned to the point where KOSPI has become a "giant self-reinforcing feedback loop" — it accelerates on the way up and accelerates on the way down.
Korea's volatility index briefly approached 100. The market has triggered six full-market circuit breakers so far in 2026. This means → since that circuit-breaker system was established in 2000, it has fired only twelve times total — this year accounts for half.
Jordi Visser, head of AI macro research at 22V Research, wrote: "Momentum volatility is now above dot-com-era levels, forcing liquidations at hedge funds with VAR limits and at retail traders chasing breakouts."
04

Why were fundamental funds hurt less?

Fundamental long-short funds — driven by human judgment rather than pure models — fell only 2.2% over the same period, retaining a strong 15.5% year-to-date return.
The key difference: fundamental managers actively trimmed AI-related long positions and pushed total leverage to the lowest decile of the past year. In plain terms = they pulled back early, reducing their bet on "AI stocks keep rising."
Yet information-technology and momentum factor losses were still the main drag on fundamental funds too. This reflects the fact that regardless of strategy style, AI and momentum sit on the same fault line in this market.
05

What to watch next?

Some investors still remember the August 2007 quant blowup, an event linked to the eventual collapse of Lehman Brothers. Current quant fund positioning has essentially reset to start-of-year levels.
Selling pressure in chip stocks and KOSPI is showing early signs of stabilization, but the key unresolved question remains: with momentum volatility above dot-com levels, will funds rapidly re-lever and chase momentum again?
This means → if funds rebuild positions quickly, the market could re-enter a self-reinforcing "rise-chase-rise" loop; if they stay cautious, volatility may gradually compress. The answer to that question will largely determine the risk trajectory of the next phase.

Content is for reference only, not financial advice.

Quant Long-Short Funds Down 3.6% in Two Weeks as Momentum Collapse Meets Korean Stock Turmoil · nashnova