Max Drawdown
Largest peak-to-trough decline in cumulative equity; a key risk metric for backtesting and live performance.
Definition
Maximum drawdown (MDD) is the largest percentage decline from a local peak in equity to a subsequent trough before a new peak. Formally: max over t of (peak_t − trough_t) / peak_t, where peak_t is the running maximum of equity up to time t.
Why it matters for backtesting
- Worst-case loss: Summarizes the largest loss an investor would have experienced.
- Capital and psychology: Drives margin requirements and investor behavior; long deep drawdowns often lead to redemptions or stop-outs.
- Strategy comparison: Strategies with similar returns but very different max drawdowns have different risk profiles.
Limitations
- Single number; does not convey drawdown duration or frequency.
- Path-dependent; same MDD can arise from one long drawdown or several shorter ones.
- Sensitive to the sample period; out-of-sample MDD can exceed backtest MDD.
Linked concepts
Calmar ratio, drawdown duration, recovery factor, volatility.