The best multi-strategy funds are not the ones with the highest headline return. They are the ones with the tightest risk discipline, and by the time you have heard of them, most are closed.
Key takeaways
- The best multi-strategy hedge funds are defined by risk and capacity discipline, not by assets under management. The biggest platform is rarely the best-run one.
- Judge a manager on drawdowns and consistency first. Millennium went seven years to March 2025 without a monthly loss above 1%.
- Pass-through fees change everything. Multi-strats kept an average of 59% of every gross dollar in 2023; check the net, never the headline.
- Access usually decides the outcome. The best platforms are closed, so the real field is narrower than any ranking suggests.
Most rankings of multi-strategy hedge funds sort by one number: assets under management. It is the wrong number to lead with. The biggest platform is not automatically the best-run one, and the manager you would most want to hold is often the one you cannot get into. Millennium ran more than $87 billion by May 2026 (Bloomberg, November 2025); Citadel deliberately handed $5 billion back to clients at the start of 2026 and began the year on $67 billion (Reuters via Yahoo Finance, December 2025). Size, in other words, moves in both directions at the top of this market, and the direction tells you more than the level.
The useful skill is how to evaluate these managers rather than how to list them. If you are weighing the pod-shop model as part of a hedge fund allocation, the useful question is not “who is largest” but “who runs risk best, keeps their fund the right size, charges a defensible price, and can you actually access them.” We will set out the criteria that separate the strong platforms from the merely large, then score the best-known names against them using real, dated figures. The uncomfortable conclusion up front: on most of these tests the best managers are closed, and that is not a coincidence.
What a multi-strategy hedge fund actually is
A multi-strategy hedge fund, or multi-manager platform, runs many semi-independent trading teams, known as pods, inside one fund. Each pod trades a relatively discrete book. A central function sits above them, setting risk limits, controlling the fund’s overall factor and market exposure, and moving capital towards teams that are working and away from those that are not. The house takes small losses quickly and recycles the freed-up risk budget elsewhere.
This is the pod model, and it now dominates institutional hedge fund capital. Global hedge fund assets stood at roughly $4.5 trillion in 2024 (HFR), and the multi-strategy platforms are among the largest single managers inside that total. The model works best in liquid markets such as equities, equity derivatives, rates, credit and foreign exchange, where a position can be cut fast and risk expressed with precision. It won share because it industrialises two problems that sink single-manager funds: risk budgeting and talent churn. A platform can lose a team, hire a team or shut a book without breaking the fund. That resilience is the product. For the wider context on where this sits among strategies, see our hedge funds guide.
So far so structural. None of it tells you which platform to rate. For that you need tests.
The five tests that separate the best managers
Here is the framework. Five criteria, in the order that matters to a capital-ready allocator.
1. Risk discipline, not headline return
The single most important thing a multi-strat sells is not a big number. It is a smooth one. The whole point of spreading capital across dozens of uncorrelated books and cutting losers fast is to produce a return that does not lurch around with the market. So the first test is how the manager behaves when things go wrong, not how high it climbs when they go right.
Judge it on drawdowns and consistency. Millennium’s flagship fell about 1.3% in March 2025, its first monthly loss of more than 1% since 2018 (Bloomberg, April 2025). Read that the right way and it is a mark in the manager’s favour: a seven-year run without a 1% down-month is the discipline you are paying for. A platform that posts a spectacular year and then gives a chunk of it back is worse, for most allocators, than one that grinds out a steadier figure. The return you keep is a function of the losses you avoid.
2. Capacity discipline: does the manager close and hand money back
This is the test almost every AUM-ranked list ignores, and it is the one that most reliably identifies a serious manager. Every strategy has a size beyond which returns decay, because the trades that generated the edge get crowded and the transaction costs of moving a larger book eat the alpha. A disciplined platform caps its own size and returns capital when it judges it cannot deploy more without hurting performance.
The best names do exactly this. Citadel returned about $5 billion of 2025 profits to investors at the start of 2026, citing limited opportunities heading into the year, and has now handed back roughly $32 billion since 2017 (Yahoo Finance, December 2025). Point72 grew towards $42 billion through 2025 but capped itself at about $41.5 billion by using strategic redemptions to hold the line (Navnoor Bawa analysis, January 2026). A manager returning money in a strong year is not admitting weakness. It is telling you it values the return per dollar over the fee on more dollars. That alignment is rare and worth a lot.
3. Pod breadth and the strength of the central engine
Two things make a platform robust: how many genuinely independent bets it runs, and how good the middle is at allocating between them. A fund with a hundred pods across equities, macro, credit, rates and quant has more shots and less reliance on any one of them. But breadth without a strong central risk function is just a large collection of correlated bets wearing a costume. The engine that sizes risk, enforces the factor limits and reallocates capital is where the real skill sits.
You cannot see a pod count from outside, so judge breadth by the spread of strategies the manager reports and by how the fund holds up when one area breaks. 2025 was a useful stress test. Several platforms took early-year losses on tariff shocks and unexplained quant drawdowns, then recovered as equities rallied in the second half (Hedgeweek, January 2026). The platforms that recovered cleanly did so because the central engine moved risk, not because any single pod saved the year.
4. Fee level: what share of the gross you actually keep
Multi-strats abandoned the old “2 and 20” model for pass-through fees, where the fund charges its running costs straight to investors, including the compensation it pays to hire and keep pod talent. Those costs are not trivial and they are not capped. Across the industry, multi-strats retained an average of 59% of every dollar of gross gains in 2023, up from 46% two years earlier (Hedgeweek, February 2025). Read that again. On average, more of the gross went to the manager than to the investor.
The figures behind that average are stark. Balyasny’s Atlas Enhanced fund made 15.2% gross in 2023, of which investors kept 2.8% after $768 million of pass-through fees. ExodusPoint charged pass-through fees of 8.4% of assets in 2023 and passed just over half its gross gains to clients. Citadel’s three largest funds levied $12.5 billion in pass-through fees between 2022 and September 2024, of which $11 billion went to employee pay (Hedgeweek, February 2025). None of that is a scandal; it is the price of the talent the model runs on. But it changes how you read a headline return. An 18% gross year and an 8% pass-through fee are not the same 18% you would keep from a cheaper fund, and the only number that matters to you is the net.
5. Access: the constraint that decides everything
The last test overrides the other four for most investors, because it is the one you cannot solve with analysis. The platforms that score best on risk, capacity and consistency are, precisely because they are disciplined about size, the hardest to get into. Citadel is closed to new investors and returning money to existing ones (Hedgeweek, February 2025). Point72 paused subscriptions to its AI fund at $1.5 billion in 2025 before reopening selectively (Navnoor Bawa analysis, January 2026). The capacity discipline that makes a manager good is the same thing that shuts the door.
That leaves a live investor with a narrower field than the rankings imply: the newer or faster-growing platforms still taking capital, allocations bought secondhand where funds permit, or exposure through a fund-of-funds or a wealth platform that already holds a slice. Millennium’s 15% stake sale in November 2025, which brought in capital via wealth platforms at UBS, Morgan Stanley, Bank of America and others (Bloomberg, November 2025), is a sign of how these firms are widening their investor base without simply opening the flagship. When you evaluate a multi-strat, evaluate whether you can hold it at all. The answer often reorders the whole list.
How the top managers score
The table below applies the framework to the best-known platforms using the most recent dated figures available. Returns are full-year 2025 net figures as reported. AUM and fee inputs carry their own dates. Treat all of it as directional: these numbers move every quarter, and pass-through fees vary by fund and share class.
| Manager | 2025 net return | AUM (dated) | Capacity discipline | Access | Notable input |
|---|---|---|---|---|---|
| Citadel (Wellington) | +10.2% | ~$67bn (Jan 2026, post-distribution) | Strong: returned ~$5bn for 2026, ~$32bn since 2017 | Closed to new capital | Pass-through wrapped into a 3 to 6% management fee |
| Millennium | +10.5% | ~$87bn (May 2026) | Steady: 7 years without a >1% down-month to March 2025 | Very hard; widening via stake sale | Fee floor ~1% of assets or 20% of gains, whichever is greater |
| Point72 | +17.5% | ~$41.5bn (capped, late 2025) | Strong: capped size, returned ~$3-5bn in early 2025 | Largely closed; selective | Beat the S&P 500’s 16.4% for 2025 |
| Balyasny | +16.7% | ~$29 to $31bn (late 2025) | Growing; taking capital | More open than the megacaps | Atlas Enhanced kept clients 2.8% of a 15.2% gross in 2023 |
| ExodusPoint | +18% | ~$12bn (Dec 2025) | Best year on record; still scaling | More open | 2023 pass-through fee ~8.4% of assets |
| Schonfeld | +12.5% | ~$10 to $15bn (est.) | Equities-heavy, process-led | Selective | Flagship +12.5% in 2025 |
Sources for the table: Hedgeweek, January 2026; Yahoo Finance, January 2026; Hedgeweek, February 2025; Bloomberg, November 2025.
Read the table across, not down, and a pattern appears. The two largest and most disciplined names, Citadel and Millennium, posted the lowest returns of the group in 2025 and are the hardest to access. The higher 2025 numbers came from the smaller, faster-growing platforms, ExodusPoint at 18%, Balyasny at 16.7%, Point72 at 17.5%, several of which beat the S&P 500’s 16.4% for the year while the giants trailed it (Yahoo Finance, January 2026). That is not evidence the giants are worse. Over a full cycle their consistency is the thing you are buying, and a single strong year for a smaller fund is a thinner track record. But it does show why sorting by AUM is a poor guide to what you would have earned, or whether you could have earned it at all.
What the framework tells you about “best”
Put the five tests together and “best” stops meaning “biggest” and starts meaning something you can actually use. The best-run multi-strats are defined by risk discipline and capacity discipline, which show up as steady returns and a willingness to hand money back. Those same qualities make them close to shut. The platforms still taking meaningful new capital tend to be earlier in their growth, which can mean stronger recent numbers and a shorter, less-tested record. And across all of them, the pass-through fee decides how much of any of it reaches you.
None of this is a recommendation to buy or avoid any manager. It is a lens for reading a market that is usually presented as a league table. When you next see the big platforms ranked by assets, run the five tests instead. The order changes, and the manager at the top of the AUM list is rarely the one the framework rates highest, or the one whose door is open.
For related reading, see our directory of the largest multi-strategy hedge funds ranked by AUM and our piece on quant fund performance through the quant winter.
FAQs
Which is the best multi-strategy hedge fund?
There is no single answer, and sorting by size gives the wrong one. On risk and capacity discipline, Citadel and Millennium are the benchmark, but both are effectively closed. On recent net return, smaller platforms such as ExodusPoint (+18% in 2025) and Point72 (+17.5%) led, though with shorter records. “Best” depends on which of the five tests matters most to you and, above all, on which funds you can access.
Why do the biggest multi-strats return capital to investors?
Because every strategy has a size beyond which returns decay. When a manager judges it cannot deploy more capital without hurting performance, handing money back protects the return per dollar for remaining investors. Citadel cited limited opportunities when returning ~$5bn for 2026.
What are pass-through fees?
Instead of a fixed “2 and 20”, multi-strats charge their running costs, mainly the pay needed to hire and keep pod talent, directly to the fund. These costs are uncapped and can be large: ExodusPoint’s 2023 pass-through fee was about 8.4% of assets. It means the gross return and the return you keep can differ sharply.
Can a normal investor access these funds?
Rarely for the top names directly, as most are closed. Realistic routes are the newer platforms still taking capital, secondary allocations where permitted, or exposure via a fund-of-funds or a wealth-management platform that already holds an allocation. Minimums and eligibility rules make these funds institutional by design, and this is general information rather than advice about your own circumstances.