Maximum Drawdown: The Deepest Peak-to-Trough Decline
Maximum drawdown is the deepest peak-to-trough equity decline. Learn to measure it, how it pairs with profit factor, and how it differs from trailing drawdown.
By Imperial Analytics
Maximum drawdown answers a question profit factor cannot. Not whether a strategy made money, but how bad the worst stretch was on the way to making it. A strategy can win far more than it loses across a window and still put an account through a decline steep enough that a trader closes it at the bottom, or a funded account breaches its floor before the recovery arrives. This post defines maximum drawdown, walks the arithmetic on an illustrative equity curve, separates it cleanly from profit factor and from a prop-firm trailing drawdown, and shows how to log it so the figure stays honest.
By Imperial Analytics
What maximum drawdown is
Maximum drawdown is the largest peak-to-trough decline an account's equity records over a measurement window. Track the running high-water mark, measure how far equity sits below it at each point, and the deepest of those declines is the maximum drawdown. It is stated in dollars or as a percent of the peak it fell from.
The high-water mark is the highest equity the account has reached so far in the window. At any moment, the current drawdown is how far equity has fallen below that running peak. Maximum drawdown is the deepest that gap ever gets. It is a single decline, not the sum of every dip, and it belongs to a specific run of trades in a specific order.
That last point is what sets the metric apart from a return total. Profit factor and expectancy add trades up, and a sum does not care what order the trades arrived in. Maximum drawdown does. It reads the equity curve as a path and reports the worst descent along that path. Two accounts that end the window at the same balance can carry very different maximum drawdowns, because one climbed steadily and the other fell into a hole first and dug out.
Stated as a percent, the decline is measured relative to the peak it fell from, not the starting balance and not the ending balance. A fall from a 13,000-dollar high-water mark to an 11,200-dollar trough is a drawdown of 1,800 dollars, or about 13.8 percent of that peak. Dollars are the natural unit for a fixed account size; percent is the unit that lets a trader compare declines across accounts of different sizes.
How to measure it from an equity curve
Walk the equity curve trade by trade. Carry the running high-water mark forward, and at each point record equity minus that peak. The most negative value over the window is the maximum drawdown. In the illustrative curve below, the deepest decline is 1,800 dollars, measured from a 13,000-dollar peak down to an 11,200-dollar trough.
The procedure is mechanical and fits in one spreadsheet column. Start with the account balance after each closed trade. In a second column, carry the highest balance seen so far, updating it only when a new high prints. In a third column, subtract the running peak from the current balance. Every value in that third column is either zero, at a new high, or negative, in a drawdown. The single most negative entry is the maximum drawdown for the window.
| Point | Equity | High-water mark | Drawdown from peak |
|---|---|---|---|
| Open | $10,000 | $10,000 | $0 |
| Peak A | $12,000 | $12,000 | $0 |
| Trough A | $11,100 | $12,000 | -$900 |
| Peak B | $13,000 | $13,000 | $0 |
| Trough B | $11,200 | $13,000 | -$1,800 |
| Recover | $12,400 | $13,000 | -$600 |
Two separate declines show up in the curve. The first bottoms 900 dollars below the 12,000-dollar peak. The second bottoms 1,800 dollars below the higher 13,000-dollar peak. Maximum drawdown reports the deeper of the two, 1,800 dollars, and ignores the shallower one. A recovery to 12,400 dollars still sits 600 dollars under the high-water mark, so the account is not out of drawdown at the end of the window even though it is well above where it opened.
Data note
The equity figures in the table, including the 13,000-dollar peak, the 11,200-dollar trough, and the resulting 1,800-dollar maximum drawdown, are illustrative and chosen to make the arithmetic legible. They are not drawn from a live account. A maximum drawdown read from a trade log is only as stable as the sample behind it; per the Imperial Analytics sample-size discipline, a per-strategy figure is treated as indicative rather than settled until the strategy has cleared the twenty-trade-per-setup floor and is shown with the sample window attached.
How maximum drawdown differs from profit factor
Profit factor is order-independent: it sums the winning and losing dollars and divides, and a sum does not change when the trades are reshuffled. Maximum drawdown is order-dependent: reshuffle the same trades and the deepest decline moves. Profit factor sizes the edge; maximum drawdown sizes the worst stretch a trader had to hold to collect it.
Take one fixed set of closed trades. Their profit factor is fixed too, because gross profit and gross loss are sums and the order of the terms cannot change a total. The same is true of expectancy, which is just the average of those same outcomes. Both metrics describe the strategy as a bag of results with the sequence thrown away.
Maximum drawdown keeps the sequence. Deal the same trades in a different order and the equity curve traces a different path, so the deepest peak-to-trough decline lands at a different value. Cluster the losers early and the drawdown is deep; spread them out and it is shallow. This is why two strategies with an identical profit factor can be very different to trade. The ratio says the edge is the same size. The drawdown says one of them made a trader sit through a hole twice as deep to reach it.
↳ Note
Profit factor tells you the edge was real. Maximum drawdown tells you what you had to survive to keep it.
That is the sense in which the two metrics complement rather than compete. One reports the quality of the return, folding win rate and payoff into a single cushion above breakeven. The other reports the risk of the path, the depth of the worst decline a trader had to hold through. A full read of a strategy needs both, because a number that looks strong on return can hide a path that few traders would actually stay in.
Why the two read together
A strong profit factor paired with a drawdown a trader cannot survive, financially or psychologically, is not a tradeable edge. A shallow drawdown with a profit factor at breakeven is not an edge at all. Read return quality and path risk as a pair, not as competing single numbers.
The pairing matters because the two failure modes are different. A strategy can fail by not having an edge, which profit factor and expectancy catch. It can also fail by having an edge that arrives through a path no account can hold, which only a drawdown measure catches. A profit factor of 1.6 across a few hundred trades describes a genuine edge. If collecting that edge means enduring a drawdown larger than the account can take without breaching a limit or the trader can take without closing at the bottom, the edge is real and untradeable at the same time.
Drawdown also sets the ceiling on position size. The larger a trader sizes, the larger every dollar swing, and the deeper the same sequence of trades cuts into equity. A strategy whose maximum drawdown is comfortable at one contract can breach a funded account at three. Sizing decisions run through the drawdown, not the profit factor, because it is the depth of the worst path, not the quality of the average trade, that ends an account.
One compact way to hold both figures together is to divide the net return over a window by the maximum drawdown over the same window. A strategy that returns three dollars for every dollar of worst-case decline is a different proposition from one that returns the same total while dipping through a decline just as large as its gains. Neither profit factor nor return alone surfaces that. Read next to edge decay, a drawdown that keeps deepening while the edge holds flat is an early sign the path is getting harder to hold even when the ratio has not moved.
Maximum drawdown versus a prop-firm trailing drawdown
A strategy's maximum drawdown is a post-hoc statistic computed from your own equity curve after the trades are done. A funded account's trailing drawdown is a rule-based floor set in advance that ratchets up with your equity and ends the account the moment it is touched. One measures a strategy; the other enforces a limit.
The two share a word and little else. Maximum drawdown is backward-looking and informational. It tells a trader how deep the worst decline was over a sample so a strategy can be judged and sized. Nothing happens the moment it is reached; it is simply recorded and read later. It is measured against the account's own high-water mark from the trader's own trades.
A prop-firm trailing drawdown is forward-looking and enforced. The provider sets a floor a fixed distance below the account's high-water mark, the floor trails equity upward as new highs print, and touching it fails the account. It is a rule, not a measurement, and its purpose is to cap the firm's risk rather than to describe the trader's strategy. The same is true of a daily loss limit, which enforces a floor inside a single session rather than across the account's life.
The distinction is worth keeping straight because the two get sized against each other. A trader picks a position size so that the strategy's realistic maximum drawdown, the informational figure, stays comfortably inside the account's trailing drawdown, the enforced floor. When the measured drawdown of a strategy approaches the enforced limit of the account, the strategy is too large for that account regardless of how strong its profit factor looks.
Logging maximum drawdown so it stays honest
Maximum drawdown is a worst-case order statistic, so it is highly sample-sensitive and can only grow as more trades are added. Log the equity curve and its running high-water mark, record the maximum drawdown with the sample window and the trade sequence that produced it, and never compare drawdowns measured over samples of different lengths as if they were equal.
The first discipline is to store the curve, not just the number. A log that keeps the running balance and the high-water mark after every trade lets a trader recompute the maximum drawdown at any time, attribute it to a specific run of dates, and see whether it came from one bad cluster or a long grind. The single figure on its own hides all of that.
The second discipline is to respect the sample. The maximum drawdown observed in a log is a lower bound on the true worst the strategy can produce, because it only reflects the sequences that happened to occur. Add more trades and the figure can only stay the same or deepen, never shrink, since a longer path has more chances to trace a worse decline. This is why the twenty-trade-per-setup floor that governs any pattern claim applies here too, and why a drawdown from forty trades cannot be read next to a drawdown from four hundred as though they measured the same thing.
The third discipline is attribution. A maximum drawdown recorded with its start and end dates, and with the trades that spanned it, tells a trader something the bare depth does not: whether the worst stretch was a handful of oversized losers, a stretch of the edge going quiet, or a sizing decision that turned an ordinary losing run into a deep one. The depth says how bad. The sequence says why. A log that keeps both is the one that turns a drawdown from a number a trader flinches at into a decision about size and strategy.
Frequently asked questions
- q: What is a good maximum drawdown for a trading strategy? a: There is no universal threshold, because a survivable drawdown depends on the account size, the enforced limits on that account, and the trader's tolerance for holding through a decline. A maximum drawdown is read against the strategy's profit factor and its sample size rather than against a fixed target.
- q: Is maximum drawdown measured in dollars or percent? a: Either. Dollars are natural for a fixed account size and map directly onto enforced limits. Percent, measured relative to the peak the decline fell from, is what lets a trader compare drawdowns across accounts of different sizes.
- q: How is maximum drawdown different from a trailing drawdown on a funded account? a: Maximum drawdown is a backward-looking statistic computed from your own equity curve to describe a strategy. A trailing drawdown is a forward-looking rule a prop firm sets in advance that trails your equity upward and fails the account when touched. One measures; the other enforces.
- q: Why does maximum drawdown grow as more trades are added? a: It is a worst-case order statistic. A longer sequence of trades has more chances to trace a deeper peak-to-trough decline, so the observed maximum can only stay the same or deepen with more trades, never shrink. The figure from a small sample understates the true worst.
- q: Can two strategies with the same profit factor have different maximum drawdowns? a: Yes. Profit factor sums trades and ignores their order, so it is unchanged by reshuffling. Maximum drawdown depends on the order, so the same set of trades dealt in a different sequence produces a different deepest decline.