Maximum drawdown measures the largest decline from a previous equity high to a later low. It is one of the most intuitive risk metrics because it answers a practical question: how much pain would an investor have experienced before the strategy recovered?
If a strategy grows from 100 to 150, then falls to 105, the drawdown from that peak is 30%. If the equity later reaches a new high, the drawdown resets. Maximum drawdown is the worst such decline in the tested period.
Drawdown matters because returns are not experienced as a smooth average. A strategy with strong annualized return may still be unusable if it suffers deep and long losses. A 50% drawdown requires a 100% gain to return to the prior high. The deeper the loss, the harder the recovery.
Drawdown should be interpreted with context. A backtest may understate future drawdowns if the data period was unusually favorable. A strategy tested only during a bull market may not reveal how it behaves in crisis regimes. For that reason, drawdown analysis should include stress periods, rolling windows, and scenario tests.
Investors also care about drawdown duration. A 20% decline that recovers in two weeks feels different from a 20% decline that takes three years to recover. Duration affects confidence, liquidity needs, and the probability that users abandon the system at the worst moment.
Position sizing should respond to drawdown expectations. If a strategy has historically produced 25% drawdowns, using leverage can quickly make the experience unacceptable. Risk control begins before the trade: it begins with deciding how much capital the strategy deserves.
Maximum drawdown is not perfect. It depends on the sample path and does not describe all future risks. But it forces a strategy conversation to move beyond return and into survival.