Maximum Drawdown - The Investor's Worst-Path Statistic

By EC Assets Research Team, Portfolio Risk · Published 16 October 2026

Maximum Drawdown: Maximum drawdown is the largest peak-to-trough decline an investment has experienced over a chosen period, expressed as a percentage. It captures the worst loss a holder would have endured had they entered at the worst possible time and exited at the worst possible time.

What Maximum Drawdown Measures

Maximum drawdown answers a question volatility does not: what is the worst loss an investor would have suffered by buying at a peak and selling at the following trough? It is computed by tracking the running maximum of the value of a portfolio and measuring how far the value falls below it, then taking the largest such fall over the period.

Two features make it useful. It is path-dependent: two strategies with the same volatility and the same average return can have very different drawdowns. And it is intuitive: a 40 percent drawdown means a 40 percent loss from the high, with no annualisation or scaling. That is why it is the risk figure most often quoted to investment committees and trustees.

How It Works

With V(t) the value at time t,

$$DD(t) = \frac{V(t)}{\max_{s \le t} V(s)} - 1, \qquad \text{MDD} = \min_t DD(t)$$

Two related statistics accompany it. Time to recover is the time from the trough until the previous peak is regained; time underwater is the whole period from the peak to that recovery.

Recovery is asymmetric. A loss L requires a gain of L / (1 - L) to recover:

Years = ln(1 / (1 - L)) / ln(1.07); the drawdown recovery calculator handles other rates.

Magdon-Ismail, Atiya, Pratap and Abu-Mostafa (2004) show that the expected maximum drawdown of a random walk grows with the length of the period observed, which is why short records understate it and why drawdowns of strategies measured over different periods are not comparable.

Drawdown also feeds ratio measures. The Calmar ratio divides annualised return by the absolute maximum drawdown, usually over three years: a strategy earning 8 percent a year with a 20 percent maximum drawdown has a Calmar ratio of 0.40. Managed futures reporting relies heavily on it.

Worked Example

A strategy starts at 100 and earns these daily returns:

The peak is set on day 2; the trough on day 7 is a hair below day 5.

The maximum drawdown is 6.37 percent. At the end of day 10 the strategy is still underwater and needs a further 2.14 percent to regain its peak.

The S&P 500 price index, using Cboe's daily closes since 1975, shows the same arithmetic at scale:

Price index without dividends; falls of 25 percent or more. Including dividends shortens the recoveries.

When It Applies (and Limitations)

Short records understate it. A two-year track record without a recession or volatility spike says little about the drawdown a strategy can produce.

Smoothed valuations hide it. Private equity, real estate and other appraisal-based assets report smoothed values, so their measured drawdowns are shallower than the losses their owners actually bear.

One number, one path. The statistic depends on a single peak and trough. Average drawdown and conditional drawdown at risk, introduced by Chekhlov, Uryasev and Zabarankin (2005), use the whole distribution of drawdowns and are more stable.

Leverage scales it, and more. Doubling leverage roughly doubles drawdowns in normal markets, but margin calls and forced sales in a falling market can make losses grow faster than leverage, and a drawdown beyond 100 percent of equity ends the strategy.

Context is needed. A 10 percent maximum drawdown is unremarkable for an equity strategy and alarming for a market-neutral one. It should be read with volatility, leverage and mandate in mind.

Why It Matters for Institutional Investors

Sizing. A trustee with a 15 percent drawdown tolerance looking at a strategy whose historical maximum drawdown was 30 percent can, on that number alone, hold no more than half of it. Grossman and Zhou (1993) show how to manage a portfolio so that it never falls below a fixed share of its previous peak.

Time matters as much as depth. For pensions paying benefits and foundations making grants, a long period underwater constrains spending for years, even if the loss itself was moderate.

Complementing forward-looking risk. Value at risk and expected shortfall estimate future losses; drawdown shows what the path actually felt like. Risk reports need both.

Controlling it. Diversification, hedging and volatility targeting all aim partly at limiting deep drawdowns, because avoiding a large loss is worth more than an equal-sized gain.

References

1. Magdon-Ismail, M., Atiya, A. F., Pratap, A., & Abu-Mostafa, Y. S. (2004). On the Maximum Drawdown of a Brownian Motion. Journal of Applied Probability, 41(1), 147-161.
2. Chekhlov, A., Uryasev, S., & Zabarankin, M. (2005). Drawdown Measure in Portfolio Optimization. International Journal of Theoretical and Applied Finance, 8(1), 13-58.
3. Grossman, S. J., & Zhou, Z. (1993). Optimal Investment Strategies for Controlling Drawdowns. Mathematical Finance, 3(3), 241-276.
4. Bacon, C. R. (2008). Practical Portfolio Performance Measurement and Attribution (2nd ed.). Wiley.