Drawdown Recovery Math: Why -50% Needs +100% to Break Even

Lose half your account and you don't need to make half of it back. You need to double what's left.

Read that again, because it's the single most expensive thing I ever learned, and I learned it the slow way. Down 50%, you need a 100% gain just to see your starting number again. Not 50%. A hundred. The loss and the recovery are not the same size, and the gap between them is where most accounts quietly die.

This is the drawdown recovery math, and it is the one calculation that should govern every risk decision you make. It doesn't care about your indicators. It doesn't care about your gut feeling or your conviction or how sure you are this trade is different. It is pure arithmetic, it is exact, and it is brutal. So let me teach it to you honestly, by hand, the way I wish someone had drilled it into me before I found out the hard way.

The asymmetry nobody warns you about

Here's the trap your brain falls into. You think of a loss and a gain as mirror images. Down 10%, up 10%, you're back to even, right?

Wrong. And it's wrong in a way that gets worse the deeper you go.

Watch what actually happens. Start with $10,000. Lose 10% and you've got $9,000. Now make 10% back — but 10% of $9,000 is only $900, not $1,000. You land at $9,900. Still down. You needed to make more than 10% to recover from a 10% loss, because you're now earning your gain on a smaller pile of money.

That's the whole secret, and it's almost insultingly simple once you see it. When you lose, your gains have to work on a shrunken base. The percentage that hurt you on the way down is calculated against a bigger number than the percentage that has to save you on the way up. Loss and recovery live on different sides of an unfair trade.

Most people never sit down and feel this. They nod at "manage your risk" and move on. But the asymmetry isn't advice. It's a law of the arithmetic, and it's coming for you whether you respect it or not.

The exact formula, worked by hand

Here's the formula. Memorize it. Tattoo it somewhere if you have to.

Gain needed to recover = DD / (1 − DD)

DD is your drawdown as a decimal — how far you've fallen from your peak. That's it. That's the whole thing. Let me run it across the full range so you can watch the trap close.

• Down 10% → 0.10 / 0.90 = +11.1% to break even.
• Down 20% → 0.20 / 0.80 = +25%.
• Down 30% → 0.30 / 0.70 = +42.9%.
• Down 40% → 0.40 / 0.60 = +66.7%.
• Down 50% → 0.50 / 0.50 = +100%. You double what's left, just to get back to zero.
• Down 60% → 0.60 / 0.40 = +150%.
• Down 70% → 0.70 / 0.30 = +233%.
• Down 80% → 0.80 / 0.20 = +300%.
• Down 90% → 0.90 / 0.10 = +900%. A 10x. Just to break even.

Look at the early numbers. Down 10%, you need 11% — annoying, but human. Down 20%, you need 25% — painful, still doable. This is the zone you can come back from on skill. A good month, a good week, you climb out.

Then look at what happens past the middle. The numbers stop adding and start exploding.

Why the curve goes vertical

Between 10% and 30%, the recovery you need creeps up gently. From 11% to 43%. Bad, but it's a slope you can walk.

Then it stops being a slope and becomes a wall.

From 50% to 90% — a span of just 40 points of drawdown — the gain you need rockets from 100% to 900%. The line doesn't bend. It goes nearly vertical. Every extra percent of loss in the deep end costs you wildly more recovery than the same percent did in the shallow end. Going from 80% down to 90% down looks like "just 10% worse." It isn't. It's the difference between needing to triple your money and needing to make ten times your money.

That's why I keep saying a shallow hole and a deep hole are different universes, not different depths of the same one. A 20% drawdown is a setback. A 50% drawdown is a crisis. A 90% drawdown is, for all practical purposes, the end — because nobody reliably makes 900% to dig out, and anyone who tells you they can is selling you the exact thing that put them in the hole.

The math is the same math the whole way down. What changes is that it turns from gravity into a black hole.

The martingale trap I survived

I know this curve in my body, not just on paper, because a bot walked me down it once.

It had a 95% win rate. The equity curve was a thing of beauty — green, green, green, a near-perfect staircase climbing the screen. What I didn't understand, because I didn't know what to look for yet, was how it was winning. Every time a trade went against it, it didn't take the loss. It doubled the next position. Down again? Double again. The strategy "recovered" almost every time, which is exactly why the win rate looked like a miracle.

But each recovery was quietly digging the hole deeper underneath the surface. The bot wasn't avoiding drawdown. It was hiding it, stacking bigger and bigger bets to paper over small losses, until one ordinary Tuesday gold just kept going one direction and the doubled-up-doubled-up-doubled-up position detonated. Months of green gone in a single afternoon. The account went from a healthy number to a rounding error, and the recovery math says it was never coming back.

That's the cruelty of martingale and grid systems. They are engineered to produce a gorgeous win rate by refusing to ever realize a small loss — which means they're engineered to convert lots of tiny survivable drawdowns into one catastrophic, unrecoverable one. A 90% drawdown needs 900% to escape. The bot didn't have a strategy for that. It had hope, and hope is not a strategy. I wrote more about exactly how this happens in why martingale EAs blow up, but the recovery curve is the why underneath the why.

Risk note: a high win rate tells you nothing about your worst day. Ask what the biggest loss looks like, always.

Why a hard drawdown ceiling is the only honest answer

So if the deep end of the curve is a black hole, the only sane move is to never get near it. Not to be clever once you're deep. To refuse to go deep in the first place.

That's a drawdown ceiling. A hard line — a maximum drawdown the account is allowed to reach, at which point everything stands down. No "just one more to win it back." No averaging into the loss. No dragging the stop lower because it'll surely bounce. A floor, decided when you're calm, that the panicked version of you at 3am cannot remove.

Why a ceiling instead of just "trading well"? Because the recovery math means the location of your worst drawdown decides everything about your future. Keep your deepest hole shallow — say, capped where recovery is still a believable single gain — and you stay in the game on skill. Let it run deep even once, and you've handed your account to a curve that now needs a miracle. The ceiling isn't timidity. It's the one rule that keeps every other rule from mattering, because a blown account doesn't get to apply its edge tomorrow.

This is the whole reason I built Axiom FX risk-first instead of chasing a flashy equity curve. Every single trade carries a hard stop, written before the trade exists. Risk per trade is capped. And the entire account sits under a max-drawdown ceiling — a line it will not cross — precisely because I've seen what's on the other side of that line. You can read the philosophy in full on how it works, and the honest numbers, worst days included, are on the results page. I lead with the worst day on purpose. Anyone hiding theirs is hiding the only part of the math that matters.

The number you should actually fear

Here's where this lands for you, whether you ever touch my software or trade entirely by hand.

When you size up a strategy, everyone stares at the wrong number. They look at the win rate, or the best month, or the total return on the marketing page. None of those decide whether you survive. The number that decides your survival is the maximum drawdown — the deepest hole the thing has ever dug, or could dig. That's the number to fear, because that's the number the recovery curve operates on.

Run any strategy's worst drawdown through DD / (1 − DD) and ask yourself one honest question: do I actually believe I can make that gain back? If a system can draw down 60% and you'd need 150% to recover, you don't have a strategy with a rough patch. You have a countdown. Judge it on the worst case, not the highlight reel. And judge your trades the same way — in R, your risk per trade as one unit, not in win rate. A system that wins less often but never lets the hole get deep will quietly outlast a 95%-win-rate machine that blows up once. I dug into that trap in the 95% win-rate trap, and it's the same lesson wearing a different mask.

So here's the takeaway I'd hand my younger self before he clicked buy on that beautiful, lethal bot. Protecting the downside isn't the boring part of trading you get to once the fun stuff is done. It is the trading. The math doesn't negotiate. Recover = DD / (1 − DD), and the curve goes vertical faster than your courage does. Keep the hole shallow, and you get to play again tomorrow. Let it go deep, and you're praying for a miracle the arithmetic already told you isn't coming.

When you're ready for a system built around that one truth — hard stop on every trade, capped risk, a drawdown ceiling, and a 30-day profit-or-refund — start here. But software or not, please: fear the right number.

Risk note: trading gold carries real risk of loss. Stops and drawdown ceilings reduce damage, they don't guarantee profit, and past performance is no promise of yours. Never risk money you can't afford to lose.

This is one trader’s opinion and education, not financial advice. Trading gold carries real risk of loss; any figures are illustrative and not a promise of results.