What maximum drawdown measures

Maximum drawdown (MDD) is the largest percentage drop from a portfolio’s peak value to its lowest point before a new peak is reached. If a fund goes from $100 million to $40 million before eventually recovering, the maximum drawdown is -60%. The formula is simple: (trough value – peak value) / peak value.

What makes drawdown different from annual return: drawdown measures the worst experience within a period, not the final result. A fund might finish the year up 20% but have experienced a -40% drawdown in the middle. The annual return looks fine. The drawdown tells you what it actually felt like to be an investor in that fund during the worst three months. For allocators who report to boards and investment committees, that lived experience matters.

In traditional hedge funds, a maximum drawdown of -20% is considered severe. A drawdown exceeding -30% often triggers redemptions that can force a fund to shut down. In crypto, those numbers are routine. A -40% drawdown in a crypto hedge fund is a bad quarter, not an existential crisis. This difference in drawdown expectations is one of the biggest barriers to institutional adoption of crypto funds. Allocators whose entire risk framework is built around -10% to -20% worst-case scenarios do not know what to do with a fund that has a -50% max drawdown in its track record.

The 2022 stress test

2022 was the worst year for crypto since the asset class went institutional. Bitcoin fell 73% from its November 2021 peak of $69,000 to a June 2022 low near $15,500. Ethereum dropped even more. And the damage went far beyond prices.

The cascade started with the Luna/TerraUSD collapse in May 2022. The algorithmic stablecoin broke its peg and went to near zero, wiping out roughly $60 billion in value. Three Arrows Capital, which had been one of the largest crypto hedge funds with approximately $10 billion in assets at its peak, had significant Luna exposure. When Luna collapsed, 3AC could not meet margin calls from its lenders (Deribit, BlockFi, Voyager, and others). The fund defaulted and entered liquidation. Total creditor claims exceeded $3.5 billion.

The contagion spread. Voyager Digital filed for bankruptcy after 3AC could not repay roughly $670 million it had borrowed. BlockFi was left with $80 million in bad loans. Celsius Network froze withdrawals. And then, in November, FTX collapsed, taking Alameda Research with it. Alameda, which functioned as a quasi-hedge fund, had borrowed heavily using the illiquid FTT token as collateral. When confidence evaporated, the collateral became worthless and both entities went bankrupt.

Multicoin Capital’s hedge fund lost 91.4% in 2022, its worst year since inception. The firm had 10% of its assets stuck on FTX and significant exposure to SOL, FTT, and SRM, all of which crashed in November. Remarkably, Multicoin remained up 1,376% net of fees since inception through 2022, which shows both the power of early-cycle gains and the scale of the drawdown.

The common thread across all of these failures was not just falling prices. It was leverage, counterparty concentration, and operational failures. Funds that held BTC and ETH in self-custody or with credible third-party custodians and avoided leverage suffered drawdowns in line with the market (bad, but recoverable). Funds that had their assets on FTX, loans to 3AC, or exposure to Luna went to zero or near it. The difference between a -60% drawdown and a -100% drawdown was operational, not strategic.

Drawdowns by strategy

Not all crypto fund strategies draw down the same way. This chart shows the range of maximum drawdowns we see across strategies in our database, including the 2022 period.

Long-only funds track the market. If Bitcoin drops 70%, a long-only Bitcoin fund drops approximately 70%. There is no hedging, no risk management beyond position sizing. The drawdowns mirror the underlying asset. For allocators with high risk tolerance and long time horizons, this is fine. For most institutions, it is not.

Long/short funds offer some protection. The short book is supposed to partially offset losses during drawdowns. In practice, the best long/short crypto funds cut their drawdowns by roughly 30-50% relative to the market. A fund that drew down -40% when Bitcoin drew down -73% is providing real value. But -40% is still -40%. It is better than the market, and it is still severe by traditional standards.

Market-neutral and arbitrage funds are in a different category. The best funds in this group kept their max drawdowns to -5% to -15% even during the 2022 crisis. Pythagoras Investments, for example, maintained a max drawdown of approximately -15% over a period when Bitcoin dropped 73%. These strategies work because their returns come from structural inefficiencies (price differences across exchanges, basis trades, funding rate arbitrage) rather than from directional crypto exposure. When the market crashes, the inefficiencies often get larger, which can actually help these strategies.

The recovery math problem

Drawdowns are asymmetric. The math of recovery is punishing, and most people underestimate how punishing until they see it laid out.

A -50% drawdown requires a +100% gain just to get back to even. At a 20% annual return rate (which is strong for any strategy), that takes nearly four years. A -90% drawdown requires a +900% gain, which at 20% per year takes over a decade. Multicoin’s -91% drawdown in 2022 would have required a 1,000%+ gain to fully recover. The fund actually managed something close to that because crypto markets had a massive rally in 2023-2024, but that outcome was exceptional and cannot be counted on.

This is why drawdown management is arguably more important than return generation for crypto funds. Avoiding the deep hole is worth more than any amount of alpha. A fund that returns 25% per year with a -20% max drawdown will compound to a higher terminal value than a fund that returns 40% per year with a -70% max drawdown, simply because the second fund spends years digging out of the hole before it can start growing again.

What causes extreme drawdowns in crypto funds

Based on the post-mortems from 2022 and earlier cycles, extreme crypto fund drawdowns fall into a few categories.

Leverage without adequate stops. Three Arrows Capital was leveraged long across multiple assets and counterparties. When prices fell, the leverage amplified the losses, margin calls came in faster than the fund could sell, and the spiral accelerated. This is the same mechanism that has killed leveraged funds in traditional markets for decades. Crypto just makes it faster because markets trade 24/7 and liquidation engines are automated.

Counterparty concentration. Funds that had significant assets on FTX lost those assets when FTX went bankrupt. Funds that had loaned money to 3AC lost that capital when 3AC defaulted. The lesson: where your assets sit matters as much as what you own. A diversified portfolio held on a single exchange is not actually diversified. See our articles on custodians and due diligence for more on this.

Illiquid position concentration. Funds with large positions in illiquid tokens (Luna, FTT, smaller DeFi governance tokens) could not exit when conditions deteriorated. The positions were too large relative to market depth. By the time they could sell, prices had already collapsed. This is the classic liquidity mismatch problem: offering quarterly or monthly redemptions on a portfolio that cannot be liquidated in a quarter.

Correlation spikes during crises. In normal markets, different crypto tokens have varying correlations. During a crisis, everything goes to correlation 1.0. Diversification within crypto stops working precisely when you need it most. A fund holding 20 different tokens still drew down 60%+ in 2022 because all 20 tokens fell together. The only true diversifier was being in strategies with no directional exposure (market-neutral) or being hedged.

Has risk management improved since 2022?

Yes, meaningfully. But it is not fixed.

The improvements are real. Most institutional crypto funds now use multiple custodians rather than concentrating on a single exchange. Counterparty risk assessment has become a standard part of the investment process. Prime brokerage services (from firms like FalconX, Hidden Road, and Copper) provide custody with exchange access, reducing the need to leave assets on exchanges. Leverage levels have come down. The industry average went from dangerously high pre-2022 to much more conservative post-2022.

But the underlying vulnerabilities have not disappeared. Crypto markets still trade 24/7, which means drawdowns can happen at 3 AM when the risk manager is asleep. Liquidity in smaller tokens is still thin. Exchanges can still fail (though the risk is lower with better regulation). And the potential for a correlated selloff that takes everything down simultaneously remains. The 2022-era risks were concentrated in a few specific entities (Luna, 3AC, FTX). The next crisis will likely involve different names, but the mechanics of leverage, liquidity, and counterparty risk are structural, not cyclical.

The October 2025 altcoin crash was a reminder. Dozens of tokens fell 40%+ in hours, overwhelming altcoin mean-reversion strategies and causing significant losses for funds dependent on altcoin order-book depth. Fundamental and altcoin-heavy strategies drew down about 23% in 2025 alone. Market-neutral funds, by contrast, were up about 14% for the year. The pattern repeats: directional exposure to volatile, illiquid assets remains the primary source of drawdown risk.

How allocators should use drawdown data

Start with your drawdown budget, not your return target. Before asking “what returns can I get from crypto funds,” ask “what is the maximum drawdown I can tolerate?” If your investment committee will pull the allocation after a -25% drawdown, then you should not be in a long/short crypto fund with a historical -50% max drawdown, no matter how good the returns look. Match the fund’s drawdown profile to your actual tolerance.

Look at drawdown duration, not just depth. A -30% drawdown that recovers in 3 months is very different from a -30% drawdown that takes 18 months to recover. Duration affects your ability to rebalance, meet capital calls, and maintain confidence with your stakeholders. Our Performance Database tracks both maximum drawdown and drawdown duration for every fund.

Compare drawdowns to the benchmark, not in absolute terms. A crypto fund that drew down -40% when Bitcoin drew down -73% captured only 55% of the downside. That is actually good performance for a directional crypto strategy. A fund that drew down -40% when Bitcoin drew down -25% captured 160% of the downside. That is terrible. Downside capture ratio (fund drawdown divided by benchmark drawdown) is often more informative than raw max drawdown.

Ask about the worst month, not just the worst drawdown. Monthly maximum loss tells you how bad a single reporting period can get. If a fund’s worst month was -18%, that is what you might see on any given month-end report. If your board meets monthly and a -18% number triggers panic, factor that into your decision.

Understand that survival is not the same as success. Multicoin lost 91% and eventually recovered because crypto had a massive rally. That recovery was market-dependent, not skill-dependent. If the market had stayed flat for three years after 2022, the fund would have remained deeply underwater. Surviving a -90% drawdown because the market bailed you out is not the same as managing risk well.

For a complete evaluation framework, see our manager evaluation guide and due diligence checklist.

FAQ

Related research

Best performing crypto funds of 2025 · Crypto fund Sharpe ratios · Annual performance review · Crypto hedge funds vs. Bitcoin · Bear market performance · Performance by strategy · Due diligence checklist · Crypto fund custodians