What is drawdown? Max drawdown, recovery math and risk of ruin
Drawdown is the most honest risk number a trading bot has, and the one beginners ignore until it ends them. It measures the peak-to-trough fall in your account — how far down you go before recovering — and it captures the lived experience of a strategy in a way average return never can. Two bots can post the same annual return while one rides a smooth curve and the other plunges 60% along the way; only drawdown tells them apart. This guide explains how max drawdown is measured, the brutal asymmetry of recovery, drawdown duration, and why drawdown — not volatility — is the risk metric that decides whether you survive.
What drawdown measures
Drawdown is the percentage drop from a peak in your equity curve to the lowest point before a new peak is made. If your account hits $10,000, falls to $7,000, then climbs again, you suffered a 30% drawdown. It is computed directly from the equity curve the backtester draws, and it is the single best proxy for the pain a strategy inflicts — the number that decides whether you can actually stick with the bot when it is losing.
Maximum drawdown
Maximum drawdown is the largest such peak-to-trough fall over the whole test period — the worst it ever got. It is the headline risk figure because it answers the only question that matters in a crisis: how bad can this get? A strategy with a 25% max drawdown and one with a 70% max drawdown are different animals even at identical returns, because the 70% one will be psychologically and financially unbearable for most traders long before it recovers.
The recovery asymmetry
A 50% drawdown does not need a 50% gain to recover — it needs 100%. The deeper the hole, the more violently the recovery math turns against you, because each loss shrinks the base the gain must work on. This asymmetry is why controlling drawdown matters far more than chasing return.
Drawdown duration
Depth is only half the story; duration — how long you spend underwater before a new high — is the half that breaks discipline. A bot can be down a modest 20% but stay there for eighteen months, and that long grind is what makes traders abandon a sound strategy at the worst moment. When you read a backtest, look at both the depth and the length of the worst drawdown.
Risk of ruin
Drawdown connects directly to risk of ruin — the probability your account falls so far it cannot recover. Heavy leverage and oversized positions both deepen drawdowns and raise ruin probability. A Monte Carlo simulation reshuffles your trade sequence to estimate the realistic range of worst-case drawdowns, because the single historical path you backtested was just one lucky (or unlucky) ordering.
Using drawdown for a bot
Set a maximum acceptable drawdown before you go live and build a kill switch that flattens positions if the bot breaches it — a core piece of risk management. Compare strategies on the Sharpe ratio and max drawdown together, never on return alone, and size positions so a realistic losing streak stays inside the drawdown you can actually stomach.
Frequently asked questions
What is drawdown in trading?
Drawdown is the percentage drop from a peak in your account equity to the lowest point before a new peak is reached. If an account goes from $10,000 to $7,000 and then recovers, that is a 30% drawdown. It captures the lived pain of a strategy far better than average return and is the best proxy for whether you can actually stick with a bot through its losing periods.
What is maximum drawdown?
Maximum drawdown is the largest peak-to-trough fall over the entire test or trading period — the worst the account ever got. It is the headline risk figure because it answers how bad things can realistically get, and two strategies with the same return but very different max drawdowns are fundamentally different in risk.
Why does a 50% drawdown need a 100% gain to recover?
Because gains and losses work on different bases. After a 50% loss your capital is halved, so you must double what remains — a 100% gain — just to get back to the start. The asymmetry worsens as drawdown deepens: an 80% loss requires a 400% gain to recover, which is why controlling drawdown matters more than chasing return.
Is drawdown a better risk measure than volatility?
For most traders, yes. Volatility treats upside and downside swings the same, but drawdown measures the actual peak-to-trough loss you have to endure, which is what tests discipline and threatens the account. Combined with drawdown duration and risk of ruin, it gives a far more honest picture of survivability than volatility alone.