Your worst drawdown so far isn’t your worst drawdown. It’s just the worst one you’ve lived through — and the sample keeps growing. Treating that number as your ceiling is how funded accounts die.
Peak-to-trough: defining max drawdown
Maximum drawdown (MDD) is the largest peak-to-trough decline in your equity curve: measured from a high-water mark down to the lowest point before a new high is made, usually expressed as a percentage of the peak. It’s the single number that captures the deepest hole the account ever fell into.
Two subtleties matter. MDD is path-dependent — the same set of trades in a different order produces a different figure, because drawdown is about the sequence, not the total. And it’s measured on equity, so if you mark open positions to market, an unrealized swing counts against you even before you close. On a prop account that distinction can be the difference between passing and breaching.
Historical max is a floor, not a ceiling
The MDD you’ve observed is a lower bound on what’s possible, never an upper bound. Add more trades and the number can only stay the same or get worse — it can never improve, because a new low can only deepen the record.
A backtest showing a 12% max drawdown over 200 trades will almost certainly show a deeper one over 2,000, simply because you’ve given the losing streaks more chances to line up. So read your historical MDD as “at least this bad” and budget for something worse. The trader who sizes for their observed worst case is sizing for a hole they’ve already climbed out of, not the one still ahead.
Expected drawdown from win rate and streak math
You can estimate drawdown from first principles instead of hoping. The probability of a losing streak of length k is roughly (1 − W) raised to the power k, so streaks that feel freakish are routine over a few hundred trades. A useful approximation for the longest losing run you should expect in N trades is:
longest streak ≈ log(N) / log(1 / (1 − W))
At a 50% win rate over a few hundred trades, that lands somewhere around seven to nine consecutive losers at some point — not as a tail event, but as the base case. Multiply that streak length by your risk-per-trade and you have a floor estimate for streak-driven drawdown. Real drawdowns run deeper still, because clusters of small winners often fail to offset the losers between them.
The recovery math: why deep holes compound
Drawdown recovery is non-linear, and the asymmetry accelerates the deeper you fall:
| Drawdown | Gain needed to recover |
|---|---|
| 10% | ~11% |
| 20% | 25% |
| 33% | ~50% |
| 50% | 100% |
| 60% | 150% |
A 10% loss barely stings — an 11% gain restores it. A 50% loss demands you double what’s left. This is the mathematical reason avoiding deep drawdowns matters more than chasing big up-months: the pain of the hole compounds against you while you’re in it. The drawdown recovery calculator makes the required climb concrete, and its most useful job is talking you out of over-sizing to escape faster.
Trailing vs static drawdown in prop rules
Prop firms don’t measure your MDD the way a backtester does. They impose a drawdown limit, and the type governs how it behaves:
- A static floor is fixed at the start and never moves, so every dollar of profit widens your cushion.
- A trailing floor ratchets up with your equity or balance and never retreats, so a give-back after a winning run can breach you while the account is still green.
Futures evaluations — such as the Topstep combine — commonly use a trailing threshold, which is exactly why traders get caught handing back an unrealized run. Which variant applies, and whether it trails balance or equity, changes your entire sizing plan. The trailing drawdown explainer walks the common forms, but confirm the exact mechanics with your firm — this is one of the numbers that varies most between products.
Sizing so drawdown stays inside your limits
The survival move is to work backward from the firm’s floor rather than forward from your ambitions. Decide the worst losing streak you must survive, multiply it by your per-trade risk, and make sure the product sits comfortably inside the firm’s limit — with margin, not right at the edge.
- Estimate your expected worst streak from your win rate, then add a buffer.
- Size so that streak leaves clear daylight above the floor, checked with the prop-firm drawdown calculator.
- On a trailing account, recompute the distance against your live equity, not the number you started the day with.
This is where continuous tracking beats end-of-day math. Shibiki builds your real equity curve from auto-journaled fills, so your true MDD and current distance-to-floor are always the live figures — not a stale estimate. The hard risk limits enforced at the broker let you set a personal floor inside the firm’s line so it trips first, turning a would-be breach into a routine stop-out. And when you run the same setup across several prop accounts, copying carries that protective margin onto every one of them.
Related: Trailing drawdown, explained · Prop-firm drawdown calculator · Drawdown recovery calculator