Recovering from a drawdown: the math you should know first
Lose 20% and you need 25% to get even. Lose 50% and you need 100%. The math of drawdown recovery, why depth compounds against you, and what to control.
Lose 20% and you need 25% to get back to even. Lose 50% and you need 100%. Lose 75% and you need 300%.
That is not a trading opinion. It is arithmetic, and it is the single most useful piece of math a new trader can memorize, because it explains why controlling the size of the hole matters more than how fast you can climb out of one.
Why the numbers are not symmetric
The asymmetry comes from the base changing underneath you. A 20% loss on $50,000 leaves $40,000. To get back to $50,000 you need $10,000, and $10,000 on a base of $40,000 is 25%, not 20%. The loss was measured against the larger number and the recovery is measured against the smaller one.
The formula is short: the gain required is L ÷ (1 − L), where L is the loss as a decimal. It is well behaved for small numbers and vicious for large ones.
Read the third column and the trap becomes obvious. The dollars you have to make back are always exactly the dollars you lost. What changes is that you have to earn them with a smaller account, at the same rate of return, which is why the percentage explodes while the dollar figure stays ordinary.
Small drawdowns are recoverable with normal performance. Large ones require performance you have never produced. Below roughly 20% the math is a nuisance; past 50% it becomes a different job, and the only reliable way to avoid needing a 100% year is to never take the 50% loss.
The second cost is time
The percentage table hides the part that actually breaks people: how long you spend underwater.
But equity curves do not recover at the average. They recover in fits, with fresh losses in the middle, and the psychological clock runs the whole time. A drawdown that takes four months to repair is four months of watching an account sit below a number it already touched. That is where people quit, usually a few weeks before the recovery. How to read that shape properly is in reading an equity curve, and the definition itself in what drawdown actually is.
Deep holes are usually dug, not suffered
A 50% drawdown almost never arrives from a strategy quietly losing. It arrives from decisions.
- Averaging into losers. Adding to a position because it is cheaper now converts a defined loss into an undefined one.
- Removing the stop. The single fastest route from a bad trade to a bad year.
- Doubling up to recover. Martingale logic feels like a plan and is a countdown, for reasons laid out in the math of martingale ruin.
Each of those is a choice made by a person, usually while already down. That is the pattern behind most account failures, and it is covered from the other direction in why most traders lose.
An edge does not make the hole shallower
A positive expectancy tells you where the equity curve drifts over thousands of trades. It says almost nothing about the worst sequence you will meet along the way. A system whose target sits at three times its risk, as this one does, breaks even at a 25% win rate and produces long losing runs as a matter of arithmetic, not malfunction.
What separates survival from ruin is the size of each bet against the capital behind it. Bet 1% and a ten-loss streak is a bruise. Bet 10% and the same streak is most of the account, and now the recovery math has moved into the ugly part of the table. That relationship, and how quickly it turns, is worked out in risk of ruin and the coin flip.
You cannot control the losing streak. You control what each loss is worth.
What to do while you are in one
- Do not increase size. Trading bigger to recover faster is the exact move that turns a normal drawdown into a permanent one.
- Do not change the rules mid-drawdown. Any tweak made while losing is fitted to the losing stretch, and you will have no way to tell whether it helped.
- Decide the abandon threshold in advance. A number and a duration, written down before you needed them, is the only defense against deciding at the worst moment.
- Keep taking every signal. Skipping trades during a drawdown reliably means missing the recovery, since you cannot know which trade starts it.
Control the drawdown, not the return
Between the system that made the same money, the one with the shallower drawdown is not slightly better. It is a different asset. It can be traded larger for the same risk, it survives a worse market, and its owner is more likely to still be running it in a year.
That is why the drawdown figure belongs next to the profit figure in any honest track record, and why a seller who publishes one without the other has told you which number is the problem. The comparison between worst case and typical case is in max drawdown vs average drawdown.
Frequently asked questions
What gain do I need to recover from a 20% drawdown?
You need 25%, because the recovery is calculated on the smaller balance that is left. A 20% loss on $50,000 leaves $40,000, and earning back the missing $10,000 from a $40,000 base is a 25% gain. The general formula is the loss divided by one minus the loss, which stays mild for small drawdowns and becomes brutal past 50%.
Is a 10% drawdown normal?
For an active futures strategy, yes. Any system that loses more often than it wins will produce clustered losses, and a drawdown in the high single digits or low double digits of the account is an ordinary event rather than a signal that something is broken. What matters is whether the depth and duration stay inside the range the strategy showed historically, which is why a published maximum drawdown figure is worth more than a published win rate.
Should I stop trading during a drawdown?
Only if a rule you wrote before the drawdown says so. Stopping in the middle usually means taking the losses and missing the recovery, because you cannot know in advance which trade begins the climb back. The defensible version is to define the depth or duration that would make you stop, in writing, while you are calm, and then follow it without renegotiating.
Does a bigger account make recovery easier?
Not by itself. Recovery is a percentage problem, so a 30% drawdown needs 42.9% back whether the account is $5,000 or $5,000,000. What a larger account changes is your ability to trade a small enough position that a bad streak stays a small percentage in the first place, which is the actual defense. Position size relative to capital does far more for survival than account size alone.
Why does the drawdown matter more than the annual return?
- Edward O. Thorp, "The Kelly Criterion in Blackjack, Sports Betting, and the Stock Market" (bet size, growth and drawdown).
- Van K. Tharp, "Trade Your Way to Financial Freedom" (position sizing and R-multiples).
- CFTC, "Customer Advisory: Be Cautious of Trading Systems and Robots."
Educational content, not investment advice. Trading futures involves substantial risk of loss and is not suitable for every investor. Any reference to Algomyth results is hypothetical, simulated over historical data, and shown in full on the performance page; past performance does not guarantee future results.
What would you have made?
Pick a start date, a capital and your markets, and run them against the record of the engines, priced at today's point values. Hypothetical, honest, and yours in thirty seconds.
More from Risk & sizing
Risk & sizingSurvivorship bias in trading: why you only ever see the winnersSurvivorship bias in trading means you only see what survived: the funds, traders and strategies that lived. How it inflates expectations, and how to check.
Risk & sizingHow much to risk per trade (and how futures change the 1% rule)How much to risk per trade? One percent of the account is the usual answer, but a fixed futures stop in dollars changes the math. Here is how.Risk & sizingThe psychology of watching a bot loseThe hardest part of automated trading is not the code, it is sitting through a normal drawdown without switching the bot off. Here is why, and how to do it.