Drawdown is the one risk number that doesn't care how you feel about your edge. It measures the single worst peak-to-trough decline your equity curve ever suffers — and the math for recovering from it is brutally asymmetric. Worked below with the exact arithmetic our calculators use.
A deep drawdown is a math problem before it's a psychology problem.Negocie no Bybit — grande liquidez, taxas baixas, bônus de boas-vindas de até US$ 30.000.Abra a conta Bybit →Maximum drawdown (MDD) is the largest percentage drop from any equity peak to the lowest point reached afterward, before a new peak is set. Not the average dip, not the final result — the single worst peak-to-trough decline anywhere in the history. A strategy can be net profitable for the year and still have suffered a 40% intra-year drawdown that would have wiped out an over-leveraged account long before the recovery happened.
Drawdown % = (Peak value − Trough value) ÷ Peak value × 100
The number that actually matters more than the drawdown itself is what it takes to undo it — and that relationship is not linear.
Because a loss and the gain that reverses it are measured against different base amounts, the percentage gain needed to recover always exceeds the percentage lost — and the gap widens fast as the drawdown deepens:
Recovery % = Drawdown % ÷ (100% − Drawdown %) × 100
| Rebaixamento | Remaining balance (from $10,000) | Ganho necessário para empatar |
|---|---|---|
| 10% | $9,000 | 11.11% |
| 20% | $8,000 | 25.00% |
| 30% | $7,000 | 42.86% |
| 40% | US$ 6.000 | 66.67% |
| 50% | US$ 5.000 | 100.00% |
| 70% | $3,000 | 233.33% |
| 80% | US$ 2.000 | 400.00% |
| 90% | US$ 1.000 | 900.00% |
A 50% drawdown doesn't need a 50% gain to fix — it needs a 100% gain, because $5,000 has to double just to get back to $10,000. Push the drawdown to 80% and the required recovery gain is 400%. This convexity is the entire reason risk managers obsess over capping drawdown early rather than relying on "the strategy will come back" — run your own numbers through the calculadora de recuperação de rebaixamento before assuming any loss is easily reversible.
A separate, equally common mistake: assuming three −20% losses in a row equal a 60% drawdown. They don't, because each loss is applied to an already-shrunken balance:
| Etapa | Balance |
|---|---|
| Start | $10,000.00 |
| After trade 1 (−20%) | $8,000.00 |
| After trade 2 (−20%) | $6,400.00 |
| After trade 3 (−20%) | $5,120.00 |
| Cumulative drawdown | 48.8%, not 60% |
Percentage losses compound multiplicatively, not additively — $10,000 × 0.8 × 0.8 × 0.8 = $5,120. This cuts both ways: it's the same reason a strategy can survive several moderate losing trades in a row without the drawdown looking as catastrophic as naive addition would suggest, but it's also why the recovery side (above) punishes depth so much harder than this intuition predicts.
Drawdown math gets far more dangerous once leverage is involved, because leverage rescales how much price movement it takes to inflict a given percentage loss on margin. The adverse move required to wipe out 100% of posted margin shrinks linearly as leverage rises:
| Aproveitar | Adverse price move that erases the position |
|---|---|
| 2x | 50.0% |
| 5x | 20.0% |
| 10x | 10.0% |
| 20x | 5.0% |
| 50x | 2.0% |
At 10x leverage, a move that would be a routine 10% drawdown on spot becomes a complete, unrecoverable account wipeout — there's no "waiting for the 11.11% recovery" because the position is liquidated first. This is the direct link between drawdown math and liquidation risk: check the exact distance for any position size against current price with the calculadora de liquidação e o max leverage for drawdown calculator before sizing up.
Return alone doesn't tell you whether a strategy's drawdown was worth tolerating — that's what the Calmar ratio measures: annualized return divided by maximum drawdown. A strategy returning 35% a year with an 18% max drawdown scores a Calmar of 1.94 — for every 1% of worst-case pain endured, it delivered roughly 1.94% of annual return. Two strategies with identical annual returns can have wildly different Calmar scores if one's max drawdown was 10% and the other's was 45%; the first is doing the same job for far less risk taken on. Pair this with the Sharpe and Sortino ratios for a fuller risk picture — Sharpe/Sortino measure volatility drag, Calmar measures the single worst scar.
Recovery % = Drawdown % ÷ (100% − Drawdown %) × 100. A 10% drawdown needs an 11.11% gain to break even, a 20% drawdown needs 25%, a 50% drawdown needs 100%, and an 80% drawdown needs 400%. The relationship is non-linear — every additional point of drawdown costs disproportionately more to undo, because losses shrink the base the next gain has to compound from.
Losing 50% of $10,000 leaves $5,000. To get back to $10,000 from $5,000 requires a 100% gain, not 50%, because percentage gains and losses are measured against different base amounts — the loss was measured against the larger starting balance, the recovery gain has to be measured against the smaller remaining balance.
No — it's smaller. Three sequential −20% moves on $10,000 compound to $10,000 × 0.8 × 0.8 × 0.8 = $5,120, a 48.8% cumulative drawdown, not 60%. Percentage losses compound multiplicatively, not additively, so sequential partial losses always produce a smaller total drawdown than naively summing the percentages would suggest.
Most professional systematic strategies target max drawdown in the 10-20% range, since recovery from anything beyond that becomes mathematically punishing (a 30% drawdown already needs a 42.9% gain to fix) and psychologically very hard to sit through live. Proprietary trading firms commonly enforce hard drawdown limits between 4% and 10% of account equity specifically because the recovery math gets worse the deeper it goes.
⚠️ Apenas educacional – não aconselhamento financeiro. A negociação de alavancagem pode perder todo o seu depósito.