Funds & ETFsFlows, fees & vehicles
Millennium, Balyasny and Other Multi-Strategy Hedge Funds Lose Money Together in October as AI Trades Unwind
Most large multi-manager hedge funds lost money in October, including Millennium, Balyasny and ExodusPoint, as AI stocks, bonds and other crowded trades fell together. A PivotalPath study finds the largest firms hold overlapping positions with roughly sixfold leverage.
From the Odd Lots podcast episode of November 11, 2026 and cited sources: The funds sold as diversifiers fell together. [3] [4] [8]
SOURCES[3] Hedgeweek: Major hedge funds suffer October losses hedgeweek.com[4] Unusual Whales: VARA Down 33%, Arini Down 26% unusualwhales.com[8] Bloomberg: Major Hedge Funds Mostly Lose in 'Nowhere to… bloomberg.com
AT Pulse research deskPublished · Updated · 5 min read · AI-assisted analysisAI-assisted, not human-edited. Methodology Corrections Who owns this site
- Original source
- Odd Lots podcast episode, independent perspective
- Brief by
- AT Pulse research desk AI-assisted; publisher and contact on the desk page
- When
- November 11, 2026 podcast episode published
- Covered
- Nov 12 edition
- Evidence
- 10 sources across 9 sites; transcript sourced
- 1The Perfect Storm That Clobbered Hedge Fundsomny.fm
- 2Transcript: The Perfect Storm That Clobbered Hedge Fundsapi.omny.fm
- 3Hedgeweek: Major hedge funds suffer October losseshedgeweek.com
- 4Unusual Whales: VARA Down 33%, Arini Down 26%unusualwhales.com
- 5HedgeCo: PivotalPath $2.2 trillion findingshedgeco.net
- 6OPB: Anthropic files for an IPOopb.org
Reach: not measured yet
The funds sold as diversifiers fell together. Millennium lost 2.3%, Balyasny 3.4% and ExodusPoint 2.4% in October. Citadel's Wellington fund (+1.7%) and Quantedge (+7.2%) were among the few that gained. Losses spread across equity, macro and credit strategies rather than staying in one area. [3] [4] [8]
PivotalPath found that the ten largest multi-strategy firms ran about $2.2 trillion of gross assets on $360 billion of net capital, roughly 6x leverage. Their US stock holdings overlap more with each other than with smaller funds' holdings. Each pod has strict loss limits: capital is halved after a 5% loss and the pod is closed at 7.5%. Those limits don't measure crowding across the whole firm. [1] [5] [7]
AI-concentrated funds took the deepest hits: VARA fell 33%, Arini 26%, Lone Pine 8.8% and Tiger Global 8.5%. On Odd Lots, the hosts asked whether most winning trades had quietly become bets on AI, held through chips, Treasuries or credit as well as through AI stocks. [1] [4]
On Nov. 5, the S&P 500 fell 4.36% and the Nasdaq fell 5.52%, with chip stocks leading the decline. Bloomberg strategist Michael Ball blames forced selling rather than new fundamentals. In his account, option dealers had to sell as prices fell, and leveraged ETFs, volatility-control funds and trend-following CTAs added to the selling. [1] [9] [10]
hedge funds bought for protection lost money in a falling market, and their holdings overlap heavily on high leverage.
Read the full assessment
Implication: allocators and risk teams may be holding one shared AI-linked risk across chips, rates and credit, which per-pod loss limits weren't built to catch.
Sources are checked; writing up the dossier now.
Executive brief
October 2026 was a bad month for the big multi-manager "pod" funds, which are sold as the diversified part of a portfolio: almost all of them lost money at the same time. Millennium fell 2.3%, Balyasny 3.4% and ExodusPoint 2.4% (Hedgeweek; Unusual Whales). A PivotalPath study found the ten largest of these funds hold heavily overlapping stocks, with about 6x leverage (HedgeCo). On Bloomberg's Odd Lots, the main question was whether nearly every winning trade had quietly become a bet on AI.
What changed and event timeline
Anthropic files for IPO
Anthropic filed its S-1 under the ticker ANT, with reported valuations of $1.5–2 trillion. This adds a large new supply of AI shares to an already crowded theme ().
PivotalPath crowding study
The ten largest multi-strategy funds ran $2.2 trillion in gross assets on $360 billion of net capital at the end of 2025. Their US equity holdings overlap more with each other than with smaller funds' holdings (;).
October returns land
Bloomberg reported losses across strategies, with "nowhere to hide." Citadel's Wellington fund (+1.7%) and Quantedge (+7.2%) were among the few gains (;).
Post-midterm rout
The S&P 500 fell 4.36% to 7,353.73, its worst day since April 2025. The Nasdaq fell 5.52% and entered a correction. Chip stocks led the decline on fears of a slower data-center buildout (;).
AI-concentrated funds tallied
Reported October returns: VARA −33%, Arini −26%, Lone Pine −8.8% and Tiger Global −8.5% ().
Odd Lots episode
Recorded Nov 10 and published Nov 11. Nishant Kumar and Michael Ball discuss crowding, pod risk rules and market mechanics ().
Capabilities and access
- Who reports returns: fund returns come from investor letters as reported by Bloomberg. They are not audited filings, and there is no position-level disclosure.
- Who can invest: demand is limited by capacity. Kumar says Millennium sought $20 billion, drew more than $30 billion of demand and took $22 billion.
Read the full section
- Who reports returns: fund returns come from investor letters as reported by Bloomberg. They are not audited filings, and there is no position-level disclosure. Kumar says the reasons Citadel and D.E. Shaw made money in October are not known (Odd Lots, ~07:00).
- Who can invest: demand is limited by capacity. Kumar says Millennium sought $20 billion, drew more than $30 billion of demand and took $22 billion. He also says D.E. Shaw, Point72 and Citadel together returned $35–40 billion to investors (~24:30).
Technical analysis for researchers and developers
- How pod risk control works: Kumar describes a common rule.
- Correlation only shows up under stress: pairwise correlations that look low in normal markets can jump when markets fall.
- Market mechanics: Ball links the Nov 5 move to forced selling. Leveraged ETFs, volatility-control funds and trend-following CTAs added to the selling (~30:00).
Read the full section
- How pod risk control works: Kumar describes a common rule. A manager who loses 5% has their capital halved, and is closed out at a total loss of 7.5% (~15:00). Every pod obeys this rule, but nothing in it measures crowding across the whole firm. PivotalPath warns that pod-level controls may not capture firm-level crowding when leveraged positions unwind (HedgeCo).
- Correlation only shows up under stress: pairwise correlations that look low in normal markets can jump when markets fall. One "AI factor" can show up through chips, Treasuries or credit.
- Market mechanics: Ball links the Nov 5 move to forced selling. Option dealers who had sold downside protection had to sell as prices fell (short gamma). Leveraged ETFs, volatility-control funds and trend-following CTAs added to the selling (~30:00).
Claims and evidence
- Millennium −2.3%: confirmed (Hedgeweek).
- VARA −33%: confirmed (Unusual Whales).
- Whale Rock: Kumar says −12%; the Nov 8 tally lists about −8% (Unusual Whales).
Read the full section
- Millennium −2.3%: confirmed (Hedgeweek).
- VARA −33%: confirmed (Unusual Whales).
- Whale Rock: Kumar says −12%; the Nov 8 tally lists about −8% (Unusual Whales).
- PivotalPath net capital: Kumar says $300–350 billion; the report says $360 billion. PivotalPath also measured correlations of 0.97 with value-neutral funds and 0.92 with equity market-neutral funds (HedgeCo).
- Anthropic IPO valuation: Tracy Alloway cites $1.4 trillion. The reviewed reporting says $1.5–2 trillion, so $1.4 trillion is not corroborated (OPB).
- Goldman's 0.2–0.3 correlation between multi-strat funds, and Rokos at roughly −5%: both come from the transcript only, and the reviewed sources do not corroborate them.
Context and prior work
Ten large multi-strategy funds now run more than triple the gross assets they had in 2018 (Bloomberg). Odd Lots has covered the same mechanics before: structured products, leveraged ETFs and earlier conversations on multi-strats. The hosts frame October as those mechanics "on full display." Kumar points to strain across the sector, citing Eisler shutting down and Walleye cutting parts of its equity trading (~23:40).
Limitations, safety and contested findings
- PivotalPath says its findings do not signal imminent disruption. It also notes that multi-year investor lockups reduce the risk of forced redemptions (HedgeCo).
- Kumar separates crowding from momentum trading: independent managers can reasonably like the same liquid stocks for their own reasons.
- Ball calls the days after Nov 5 mostly mechanical rather than a change in fundamentals. That is one strategist's view.
Read the full section
- PivotalPath says its findings do not signal imminent disruption. It also notes that multi-year investor lockups reduce the risk of forced redemptions (HedgeCo).
- Kumar separates crowding from momentum trading: independent managers can reasonably like the same liquid stocks for their own reasons.
- Ball calls the days after Nov 5 mostly mechanical rather than a change in fundamentals. He says AI compute demand and order backlogs still look strong. That is one strategist's view.
- Odd Lots is commentary, and its figures sometimes differ from the written reporting, as noted above.
Business and practitioner implications
- Allocators: a 2.3% loss matters because these funds are bought as a hedge.
- AI infrastructure firms: hedge-fund positioning now moves the cost of capital for neoclouds and chipmakers.
- Risk teams: treat exposure to the AI theme as one shared risk factor across chips, rates and credit, rather than as separate bets.
Read the full section
- Allocators: a 2.3% loss matters because these funds are bought as a hedge. Losing in a down month undercuts the reason pensions and sovereign funds pay for them.
- AI infrastructure firms: hedge-fund positioning now moves the cost of capital for neoclouds and chipmakers. Ball expects scrutiny to shift to data-center projects for 2028 that still face state-level opposition.
- Risk teams: treat exposure to the AI theme as one shared risk factor across chips, rates and credit, rather than as separate bets.
The source trail.
Sources (10)
The Perfect Storm That Clobbered Hedge Funds
Transcript retrieved via published_transcript; language en. Timestamped text, not direct audiovisual review. Source: https://api.omny.fm/orgs/e73c998e-6e60-432f-8610-ae210140c5b1/clips/d50c15fb-6c5a-4836-89a2-b4fe0174107c/transcript?format=TextWithTimestamps&t=1794350284. Automatic captions/transcription may contain errors.
omny.fmTranscript: The Perfect Storm That Clobbered Hedge Funds
published_transcript
api.omny.fm