What is Drawdown in Trading? The Most Important Number You Ignore
TL;DR: Drawdown is the percentage drop from your account's highest equity point to its lowest before a new high. It measures pain, not just performance. A 50% drawdown requires a 100% gain to recover — which is why drawdown limits matter more than peak returns.
The simple definition
Imagine your account balance over time. It goes up, it goes down, it makes new highs, sometimes it falls hard.
Drawdown is the gap between any equity peak and the trough that follows it, expressed as a percentage of the peak.
If your account reaches $10,000, then drops to $8,500 before rising again, you've experienced a 15% drawdown ($1,500 ÷ $10,000).
Maximum drawdown is the biggest such gap across your entire trading history. It's the number that quantifies the worst stretch you've actually lived through.
Why drawdown matters more than total return
Two strategies, both end the year up 30%:
- Strategy A: Smooth equity curve, max drawdown 8%.
- Strategy B: Volatile equity curve, max drawdown 35%.
Strategy A is dramatically better, even though the headline returns match. Why?
-
Recovery math. A 35% drawdown requires a 54% gain to break even. A 50% drawdown requires 100%. A 75% drawdown requires 300%. Losses compound asymmetrically — recovery isn't linear.
-
Real-world response. You won't trade Strategy B the same way you trade Strategy A. After a 30% loss you'll second-guess every entry. You'll cut winners early and hold losers in hope. The strategy's edge depends on you executing it consistently — drawdown destroys that consistency.
-
Risk of ruin. A 50% drawdown isn't twice as bad as 25%; it's several times worse statistically. The deeper the drawdown, the higher the probability of further losses pushing you into account-ending territory.
This is why every serious risk metric (Sharpe, Sortino, Calmar) factors in some measure of drawdown or downside volatility, not just average return.
How to calculate drawdown
For any account or equity curve:
- Walk forward through the equity series.
- Track the running maximum equity value (the highest peak seen so far).
- At each point, compute:
drawdown_pct = (peak − current) ÷ peak × 100. - The maximum drawdown is the largest value of drawdown_pct across the whole history.
In code, this is a 4-line loop. Most trading platforms and backtest reports compute it automatically.
Worked example: - Day 1: equity $10,000 (peak = $10,000, dd = 0%) - Day 2: equity $10,500 (peak = $10,500, dd = 0%) - Day 3: equity $9,800 (peak = $10,500, dd = 6.7%) - Day 4: equity $9,400 (peak = $10,500, dd = 10.5%) - Day 5: equity $10,800 (peak = $10,800, dd = 0%)
Max drawdown over this period: 10.5%.
Realized vs unrealized drawdown
Two flavours, often confused:
- Realized drawdown is calculated on closed trades only. It's the equity-curve drop based on settled wins and losses.
- Unrealized (floating) drawdown includes the open position's current PnL. It can be much larger if you're holding a deeply losing trade you haven't closed.
Prop firms typically measure unrealized drawdown — your worst-ever intraday equity, including open positions. This is stricter and is why prop accounts often blow up on a single big losing trade that "would have come back" if you'd held longer. The rule check fires at the trough.
Make sure you know which definition your strategy report uses. A "5% max drawdown" on closed trades can hide a 15% intraday spike that would have busted a prop account.
What's a reasonable drawdown?
Depends on the strategy style, but rough industry norms for retail:
- Conservative (mean-reversion, low frequency): max DD 5–10%
- Moderate (trend-following, swing): 10–20%
- Aggressive (high-leverage scalping, news trading): 20–40%
Prop firm limits give another benchmark: - Typical "challenge" max DD limit: 8–12% - Daily DD limit: 4–5%
If your backtest shows a 35% max DD, you should not run that strategy on a prop firm. You'll likely bust the limit within months. The strategy may be profitable; it's just not prop-compatible without significant position-size reduction.
The 1% rule, applied to drawdown
Risk 1% per trade. How does that translate to drawdown?
Statistically, even with a 50% win rate and a 1:1 reward-to-risk, you can hit losing streaks of 10+ trades in a row in a long history. That's 10% drawdown from a streak alone — no strategy edge needed, just bad luck.
For a real trading career (thousands of trades), you should expect at some point:
- 5-trade losing streak: virtually guaranteed
- 10-trade losing streak: likely within 1–2 years
- 15-trade losing streak: possible over a decade
If your position sizing assumes you'll never see a 10-trade losing streak, your assumption is wrong. Size so the expected drawdown from a normal streak is survivable both financially and psychologically.
Drawdown recovery time
A less-discussed metric: how long does it take to climb back from a drawdown?
A strategy that loses 10% and recovers in 3 weeks is fundamentally different from one that loses 10% and takes 18 months to recover. The annual-return numbers might be identical, but the latter is much harder to trade through.
Recovery time also matters for compounding. While you're in a drawdown, you're not making new equity highs — you're just clawing back to old ones. Long drawdowns kill the multiplier effect of compounding.
Drawdown red flags in backtests
When evaluating a strategy's backtest report:
- DD looks too small for the timeframe. If a swing strategy shows 2% max DD over 5 years, it's probably overfit. Real systems have rough patches.
- One huge DD followed by smooth recovery. Check whether the recovery is from real new edges or from a parameter optimisation that happens to fit the recovery period.
- DD concentrated in a single regime. If 80% of drawdown happens in 2008-style volatility, your strategy is vulnerable to volatility regimes — manage exposure accordingly.
- Max DD timestamps near the end of the test. A strategy with worsening DD over time is degrading. Don't assume "it'll mean-revert back."
The psychology layer
This is the part most posts skip: drawdowns are emotionally brutal in ways the numbers don't capture.
A backtest of a 20% drawdown is just an equity-curve dip on a chart. Live, it's three months of every trade going wrong, watching friends post wins on Twitter, questioning whether you ever had an edge, and being tempted to over-leverage to "make it back fast" (which makes everything worse).
You won't trade your strategy the same way during a drawdown. That's why the right time to design your drawdown tolerance is before you're in one — not in the middle.
A reasonable test: "If I lost X% in the next 30 days, would I still trade this strategy the same way the next day?" If the honest answer is no, your position sizing is too aggressive.
FAQ
Is drawdown the same as a loss? No. A single losing trade is a loss; drawdown is the cumulative peak-to-trough drop across multiple trades or days. You can have a losing trade and zero drawdown if your account is still above its previous high.
What's the difference between max DD and average DD? Max DD is the single worst peak-to-trough event. Average DD is the mean across all such events. A strategy with low average DD but a few large max DDs has fat-tail risk — most periods are smooth, but the bad ones are very bad.
Can drawdown be calculated on dollar terms instead of %? Yes, and prop firms usually use dollar amounts. A "$1,000 max DD on a $10,000 account" is a 10% rule. Percentage is more universal across account sizes; dollars are what your rule actually triggers on.
Does compounding affect drawdown? Yes. If you risk 1% of current equity per trade, your absolute risk shrinks during drawdowns (because equity is lower). This is the right way to size — it produces a self-limiting drawdown. Risking 1% of initial equity is fixed-fractional and grows DD during bad streaks.
How is drawdown different from volatility? Volatility is the standard deviation of returns — measures spread in both directions. Drawdown only measures downside. A strategy can have high volatility with low DD (volatility from large wins) or low volatility with high DD (slow grinding losses). Both matter; they answer different questions.
The bottom line
Total return tells you what the strategy made. Drawdown tells you what it cost — in equity, in time, and in nerves. Optimise for drawdown first, return second. The strategies that survive long-term are rarely the ones with the highest peaks; they're the ones with the shallowest valleys.