Drawdown, Max DD Duration, and Recovery

Drawdown is the gap between current equity and the highest equity ever reached. It is the only risk metric that captures what investors actually feel — the experience of watching your account shrink for an extended period. Two strategies with identical Sharpe can have wildly different drawdown profiles, and the difference is often what determines whether you can actually trade them.

Three Numbers, Not One

"Max drawdown" is the single most-quoted risk number, and on its own it is misleading. The full picture requires three numbers:

Drawdown Depth

How far below peak equity did you go? A 20% drawdown is uncomfortable; a 40% drawdown is for most people unbearable.

Drawdown Duration

How long did you stay below the previous peak before reaching new highs? A strategy with a 20% drawdown that recovers in 6 months is tradable. The same 20% drawdown lasting 4 years is not — you will quit before the recovery.

Drawdown Frequency

How often does the strategy enter a meaningful drawdown? A strategy that has one painful drawdown every 5 years is psychologically different from one with three painful drawdowns per year, even if the maxima are similar.

The Tradability Test

The question is not "can the strategy survive a 30% drawdown?" — historically, every interesting strategy has done so. The question is: can you survive trading it through a 30% drawdown? Most people overestimate their tolerance by a factor of 2.

Why Backtest Max DD Underestimates Live Max DD

  • Limited sample. A 5-year backtest contains 5 years of bad luck. Your future trading life contains many more. Expect to see drawdowns somewhat worse than the historical maximum.
  • Regime shifts. The conditions that produced your backtest will not exactly recur. New regimes can produce drawdowns the historical sample never saw.
  • Cost and slippage. Live costs are usually higher than backtest assumptions, eating into recovery.
  • Behavioral leakage. The backtest assumes you stayed disciplined through every drawdown. Your live performance will reflect any moments you didn't.

A common rule of thumb: multiply backtest max DD by 1.5 to estimate plausible live max DD. A backtest showing 20% max DD is a strategy you should be prepared to trade through 30% drawdowns.

The Calmar Ratio

Calmar = annualized return / max drawdown. Unlike Sharpe, it directly trades off return against tail-risk. A Calmar of 1.0+ is excellent; 0.5 is decent; below 0.3 means the drawdown profile is hard to live with.

Calmar is particularly useful when comparing strategies of similar Sharpe — the higher-Calmar strategy is generally easier to trade.

How to Use Drawdown Information

  1. Look at duration as much as depth. The depth tells you how scary it gets; the duration tells you how long you have to be scared.
  2. Stress-test your sizing. If the strategy has historical max DD of 20%, ask: at your current sizing, what dollar amount does a 30% DD represent? Can you live with that number for 18 months?
  3. Compare per-strategy DDs in a portfolio. Combining two strategies with low DD correlation can produce a portfolio with much lower combined DD than either alone — the strongest argument for diversification.
  4. Use the worst end of the bootstrap CI for max DD. The historical max is a sample minimum; the bootstrap upper bound is a more honest forward-looking worst-case.

The Bottom Line

Drawdown is the cost of admission. Every real strategy has them. The question is not whether you will hit a drawdown — you will — but whether the depth, duration, and frequency are within what you can survive operationally and emotionally. Pick strategies whose drawdown profile you can live through, then size them so that the drawdown amounts in dollars are bearable.

Further Reading

Foundational papers

  • Magdon-Ismail, M. & Atiya, A. F. (2004). Maximum Drawdown. Risk Magazine, 17(10), 99–102.

Textbook references

  • Chan, E. P. (2013). Algorithmic Trading: Winning Strategies and Their Rationale. Wiley.
  • López de Prado, M. (2018). Advances in Financial Machine Learning. Wiley.

Related QuanterLab articles

Try it in QuanterLab

On any backtest result, look at the Max DD Duration metric — not just the depth. A strategy can recover a 20% drawdown in 6 months or in 4 years; only one of those is psychologically tradable.

Back to QuanterLab
Report
Loading report...
Article
Loading article...