Maximum drawdown: from the equity peak or from capital?
Two ways of expressing drawdown as a percentage produce numbers that differ several times over. Both appear in reports, and almost nobody states which one they used.
Maximum drawdown in money terms is not controversial: the deepest fall from a previous high to the following low. Nothing to argue about — until you need percentages. And you always need percentages, because without them two accounts of different sizes cannot be compared.
Two denominators
From the peak. Divide the fall by equity at the high it started from. Answers: how much did I lose relative to my best moment.
From capital. Divide by capital — starting or current. Answers: how much of my money was exposed.
The difference is not cosmetic. An account grows from 100,000 to 300,000, then falls to 200,000. From the peak that is 33%. From starting capital it is 100%. One drawdown, a threefold spread, and both numbers are honest.
Why this matters more than it looks
Drawdown is not decoration. It feeds position sizing, stress tests and the margin buffer you need to survive a bad run. A denominator that is off by a factor of three puts every one of those conclusions off by a factor of three.
Worse: hardly any report says which method it used. You see "maximum drawdown 18%" with no way to know eighteen percent of what.
Our own case
In MaxProfit percentage drawdown was computed from the equity peak while neighbouring metrics computed from capital. Strictly speaking nothing was wrong — but a single report disagreed with itself in a way no reader could detect. We moved it to capital in 6.12.
The same release fixed the drawdown curve, which on some samples drew a vertical picket fence instead of a curve, and return, which was not being measured against capital at the start of the period.
What else hides behind one number
Closed trades or equity. Drawdown computed from closed results is blind to what happened inside an open position. Real equity drawdown is almost always deeper — sometimes twice as deep.
Duration. Down 20% and back within a week is not the same experience as down 20% and back after nine months. The second one breaks people, though the number is identical. Track depth and time under water.
One realisation among many. Your historical drawdown is a single drawn ticket. Reshuffle the same trades and the maximum changes. That is the point of simulation: it shows not your one drawdown but the distribution of drawdowns you could have had.
What to do about it
- Find out what your tool divides by. If it is not documented, verify by hand on a simple example.
- Pick one method and stay with it. Comparing peak-based with capital-based drawdown is meaningless.
- Use equity drawdown, not just closed-trade drawdown, if you hold positions longer than a day.
- Record recovery time alongside depth.
And never treat your historical drawdown as the worst case. It is merely the one that already happened.
Read next
Profit factor lies on small samples
Thirty trades and a profit factor of 2.1 mean nothing. How many trades the metric actually needs — and what to look at until you have them.
MFE and MAE: what you left on the table and what you sat through
The result of a trade tells you how it ended. MFE and MAE tell you what happened on the way — and that is usually where the fixable part is.
Expected value per trade: why three calculators give three different answers
The formula is short and everybody knows it, yet everybody gets a different number. Here is where the figures actually diverge — including the six months we showed three of them ourselves.