Drawdown Maths: Why a 25% Loss Needs a 33% Gain to Recover

Two traders finish the month down the same amount of money. One is calm. The other is in trouble. The difference isn't the loss — it's the drawdown, and specifically how far each account had to fall from its own high-water mark to get there.

Drawdown is probably the least intuitive number on a trading statement. Most platforms show it, few traders use it properly, and almost everyone underestimates what it costs to climb back out. The maths is not complicated. It is just asymmetric, and that asymmetry is the whole lesson.

Drawdown is measured from the peak, not from your deposit

The common misreading is to compare the current balance against the original deposit. That's a profit-and-loss figure, not a drawdown.

Drawdown measures the decline from the highest equity your account has ever reached — the high-water mark — down to the lowest point before a new high is made. If you deposit 10,000, run the account up to 14,000, then fall back to 11,200, you are still up 12% on deposit. But you are in a 20% drawdown, because the measurement starts at 14,000.

That distinction matters because risk compounds from where you are, not from where you started. The account that has to recover 2,800 to make a new high is working with 11,200 of capital to do it, not 14,000. Every percentage return it generates from here is calculated on a smaller base.

It's also worth separating two versions of the number:

• Balance drawdown uses closed trades only.
• Equity drawdown includes floating profit and loss on open positions.

Equity drawdown is the honest one. It is also the number your margin level is calculated against, which is why a position that is "only on paper" can still force a decision.

The recovery curve bends against you

Here is the part that surprises people. A loss and its recovery are not the same percentage.

Lose 10%, and you need 11.1% to get back. Lose 20%, you need 25%. Lose 33%, you need 50%. Lose 50%, you need 100%. The formula is straightforward:

Required gain = drawdown ÷ (1 − drawdown)

The reason is the shrinking base. A 25% drawdown on a 10,000 account leaves 7,500. To get 7,500 back to 10,000 you need 2,500 — and 2,500 on a base of 7,500 is 33.3%, not 25%.

The curve is gentle at first and then steepens sharply. Below roughly 20%, recovery looks like ordinary trading. Past 40%, it starts to require returns most strategies don't produce consistently. That is why experienced risk managers care far more about limiting drawdown than about maximising the good months. The downside is the part that changes the arithmetic of everything that follows.

A worked example: same win rate, different risk per trade

All figures below are illustrative — they are not forecasts, historical results, or any indication of what an account would do.

Two hypothetical traders run the identical strategy on a 10,000 account. It wins 50% of the time, with winners at 1.5R and losers at 1R. The only difference is risk per trade.

Trader A risks 1% per trade. A run of six consecutive losses — uncomfortable but entirely normal at a 50% win rate — costs roughly 5.9% of equity. Account sits near 9,415. The required gain to make a new high is about 6.2%. At 1% risk and 1.5R winners, that's a handful of good trades.

Trader B risks 5% per trade. The same six-loss run costs roughly 26.5%. Account sits near 7,351. Required gain to make a new high: about 36%. At 5% risk, that's attainable — but the next six-loss run starts from 7,351, and a second one would take the account to roughly 5,400, now needing 85% to recover.

Same strategy. Same sequence of trades. The position sizing alone decided whether the loss run was an inconvenience or a structural problem.

Notice also what happens to behaviour. Trader B is far more likely to widen a stop, double a size, or abandon the strategy entirely — not because the edge changed, but because the recovery number became psychologically unmanageable. Drawdown damages process before it damages capital.

Drawdown has a time axis most traders ignore

There is a second dimension to the number: how long you spend below the high-water mark. This is sometimes called time underwater or drawdown duration.

A 12% drawdown that resolves in nine trading days is a different experience from a 12% drawdown that lasts four months. The capital figure is identical. The cost in patience, opportunity, and discipline is not.

This is where the economic calendar becomes relevant to risk sizing rather than just to trade selection. This week carries ISM Services PMI on Monday 5 October at 14:00 GMT+3 (forecast 55.7 against 55.4 prior), followed by a run of Fed speakers — Williams at 13:05 GMT+3 and Bowman at 14:45 GMT+3 on Tuesday 6 October, with Logan at 23:00 GMT+3 the same day. Clustered scheduled events tend to compress several days of range into a few hours.

If you are already in a meaningful drawdown, the relevant question isn't whether a print will move your pair. It's whether your current position size still makes sense given the equity you have now, rather than the equity you had at the peak. Risk percentages should be recalculated against current equity — not left on the size that was appropriate before the decline.

The takeaway: set your floor before you need it

Pick a maximum drawdown you will accept — a figure, written down, before the month starts. Many traders land somewhere between 10% and 20%. Then size every position so that a realistic losing streak cannot breach it.

Work backwards: if you can tolerate 15% and you expect loss runs of up to eight trades, roughly 1.5% risk per trade keeps you inside the line with room to spare.

At GCC Brokers, every position is executed automatically — no requotes, no filtering by profitability, symmetric slippage — so your fills reflect market conditions rather than account performance. What your fills can't do is size your trades for you. That part stays with you, and it's the input that decides which side of the recovery curve you end up on.

If you want to check where you currently stand, pull your equity high-water mark from your statement and run the formula. It takes a minute, and the number is usually more instructive than the P&L sitting next to it.