Drawdown is the fall of an asset or portfolio from its last peak to the subsequent trough, expressed as a percentage. It answers investing's most visceral question: "how much would I have been down at the worst moment?".
Drawdown = (Current value − Previous peak) / Previous peak × 100The maximum drawdown is the worst of those falls over a period: the figure that best anticipates how your portfolio will feel in a crisis.
Recovering from a fall requires a rise larger than the fall itself:
| Fall | Rise needed to recover |
|---|---|
| −10% | +11% |
| −20% |
| +25% |
| −30% | +43% |
| −50% | +100% |
Losing 50% requires doubling your capital just to get back to the start. This asymmetry explains why limiting falls matters more than chasing rises — and why asset allocation puts so much effort into capping the plausible drawdown.
The lesson is double: deep falls are normal (not system failures), and their duration matters as much as their depth — there are drawdowns of months and drawdowns of a decade.
Volatility measures the average swing; drawdown, the worst actual blow. To decide how much risk you can bear, drawdown is more useful: nobody abandons their plan because the standard deviation is 18%, but plenty of people sell when their portfolio is 40% down. Ask yourself what maximum drawdown you would sit through without selling and build the portfolio from there.
For dividend investors there is a kind nuance: the income falls much less than prices. In 2008, S&P 500 dividends fell far less than share prices, and whoever lived off the income did not need to sell at the bottom.
The largest fall from a peak to the following trough over a given period. It is the most intuitive risk metric: it tells you how much you would have lost buying at the worst moment.
It depends on its depth and the era: the 2020 COVID fall recovered in months; 2008 took about 5 years; the dot-com Nasdaq, 15. That is why money invested in equities needs a long horizon.
With real diversification across asset classes: bonds and cash cushion equity falls. The cost is giving up part of the expected return — calibrating that trade-off is precisely what asset allocation does.