Maximum drawdown
Returns tell you how a strategy performed; maximum drawdown tells you what it cost to hold on. For many traders it is the single most important risk number.
- Needs +11% back
- −10%
- Needs +33% back
- −25%
- Needs +100% back
- −50%
- losses compound asymmetrically
Maximum drawdown, explained.
Maximum drawdown (MDD) is the largest peak-to-trough decline in account value over a period, expressed as a percentage of the peak. If an account grows to 10,000, falls to 6,500, and later recovers, the maximum drawdown is 35% — the worst experience of anyone riding the whole curve.
Drawdown is measured continuously: every new equity high sets a fresh peak, and every subsequent dip is measured against it. Maximum drawdown is simply the deepest of those dips. Two related numbers complete the picture — drawdown duration (how long the account stayed below the peak) and time to recovery.
The metric's power is psychological and mathematical at once. Psychologically, drawdowns are when traders abandon sound strategies at the worst moment. Mathematically, losses compound asymmetrically: a 50% drawdown needs a 100% gain to break even. That asymmetry is why risk management obsesses over limiting drawdown rather than maximizing return.
From idea to a running bot.
Reading drawdown from a backtest well takes more than the headline number.
Locate the worst stretch
Find when the maximum drawdown happened and what the market was doing. A strategy that only breaks in one specific regime is telling you something a summary statistic hides.
Check depth AND duration
A sharp 20% dip recovered in a week is a different experience from a slow 20% grind lasting six months. Duration is often the harder psychological test.
Stress beyond the sample
History showed one path. Monte Carlo simulation reshuffles the same trades thousands of times to estimate how bad the drawdown could plausibly have been — usually worse than the single historical number.
Built for the way you trade.
Drawdown numbers deserve attention before a single trade is placed.
Position sizers
The tolerable account-level drawdown works backwards into position sizes and leverage. Deciding it in advance — calmly — beats discovering it live.
Strategy evaluators
Between two equally profitable backtests, the one with the shallower, shorter drawdowns is almost always the one a human can actually stick with.
VolatiCloud backtesters
Every VolatiCloud backtest reports maximum drawdown with the equity curve, and Monte Carlo simulation estimates the distribution around it.
- Deepest peak-to-trough fall, as a percentage of the peak
- A 50% drawdown needs a 100% gain to recover
- Depth and duration both matter
- The denominator of the Calmar ratio
- Reported in every VolatiCloud backtest; stressed by Monte Carlo
Frequently asked questions.
What is an acceptable maximum drawdown?
It depends entirely on your risk tolerance and capital. Many systematic traders aim to keep expected drawdowns in the 10-25% range by sizing positions accordingly. The honest test: pick the number you could endure without abandoning the strategy, then size so history stays inside it.
Why does maximum drawdown matter more than volatility for some traders?
Because it maps onto lived experience. Volatility is an abstract dispersion number; drawdown is the actual worst stretch — how deep the account fell and how long it stayed down. Strategies are abandoned in drawdowns, not in volatility.
Is a backtest's maximum drawdown the worst I can expect?
No — it is the worst that happened on one historical path. Live trading can exceed it. Monte Carlo simulation, which reshuffles trade order many times, typically shows meaningful odds of deeper drawdowns than the single backtest number.
How do bots limit drawdown?
Through per-trade stop-losses, position sizing, and exposure limits — rules a bot enforces without hesitation. VolatiCloud strategies configure stop-loss and risk settings per bot, and backtests show the resulting historical drawdown before anything goes live.
Related capabilities.
Ship your first live bot this afternoon.
Connect an exchange, build a strategy in the visual builder, backtest it on real data, and deploy. Start a 7-day Pro trial — no credit card required.
No credit card required · Cancel any time