Maximum Drawdown
Education only · our voice · free public data
Definition
The largest peak-to-trough percentage decline in the equity curve over a period, measured from a running high-water mark to the subsequent lowest point before a new high is made.
How to read it
Max drawdown is the pain metric: it answers 'how bad did it get?' rather than 'how much on average?'. It drives real-world survivability because investors redeem and risk limits trip on drawdown, not on volatility. Pair it with the drawdown DURATION (time underwater) and with return via the Calmar ratio (annual return / max DD). Unlike volatility, max drawdown is path-dependent and its expected magnitude grows with the length of the track record, so a longer backtest almost always shows a deeper max DD - never compare max DDs across different sample lengths.
How practitioners use it
Used as context among multiple indicators — never as a standalone signal to act.
Less common professional uses
For a Brownian-motion equity curve the EXPECTED maximum drawdown scales with volatility x sqrt(time) and grows without bound as the horizon lengthens; a deeper max DD in a longer sample can be pure sampling artifact, not deterioration. Max drawdown is extremely sensitive to a single outlier path, so its out-of-sample value is a poor predictor of its future self; conditional metrics (drawdown-at-risk, expected time-to-recovery) are more stable. Negative-skew strategies (short vol, carry) hide their true max DD until a tail event; a benign backtest max DD from a calm sample understates the structural left tail - stress the drawdown against a crisis regime, not just the sample.
Sources & provenance
Magdon-Ismail & Atiya (2004), 'Maximum Drawdown', Risk Magazine; Bailey & Lopez de Prado (2015) on drawdown-based stopping
This page is educational content published by Pachira Aquatica Global LLC. It is not investment advice and not a recommendation.