Position Sizing That Survives a Losing Streak

A perfectly profitable system will still hand you six, eight, ten losers in a row. Not might — will. The only question that matters is whether your position sizing lets you still be trading when the streak ends.

The losing streak is a feature, not a bug

Traders treat a run of losses like a system failure. It usually isn’t. If your win rate is, say, around half, then multi-loss streaks aren’t just possible — they’re mathematically expected across a normal sample of trades. Flip a coin two hundred times and you’ll see runs of heads that feel impossible until you count them. Your trades are the same. The streak isn’t the market punishing you; it’s variance being variance.

This is exactly where the majority of traders lose — not because their edge died, but because they sized as if the streak would never come, and it took them out before their positive expectancy could play through. The entry was fine. The math on the size behind it was reckless.

Fixed fractional: the survivor’s default

The most robust answer for funded accounts is boring on purpose: risk the same small fraction of your account on every trade. Same percentage, every time, regardless of conviction, regardless of the last result.

Why it survives streaks:

Losses shrink your risk automatically. Risk a fixed fraction, and every loser makes the next trade’s dollar risk a little smaller. The account decays gently instead of falling off a cliff.
It removes the emotional dial. No “I’m sure about this one” up-sizing, no “I need to make it back” doubling. Same math, every click.
It’s honest about ruin. With a fixed small fraction, a realistic losing streak is a drawdown you can recover from — not an account-ending event.

Translate your chosen fraction into an exact size with a position size calculator every single trade. Eyeballing lot size is how “small consistent risk” quietly becomes “one oversized trade that undid the month.”

Why over-sizing is a math problem, not a nerve problem

Here’s the part that feels counterintuitive. Recovering from a drawdown is asymmetric and it gets brutal fast. Lose a modest slice of your account and you need a slightly larger gain to get back to even. Lose a big slice and the required recovery balloons — the deeper the hole, the steeper the climb, non-linearly. A drawdown-recovery calculator makes this vivid in about ten seconds, and it’s the single most sobering number in trading.

So over-sizing doesn’t just risk a bad day. It risks digging a hole your edge mathematically cannot climb out of before the funded account’s rules end the story. On a prop account this is doubly true, because a trailing drawdown can lock in your losses at the peak and shrink the room you have to recover in. Confirm your firm’s exact drawdown mechanics — static versus trailing, and where it’s measured from — because it changes how much streak your size can absorb.

Anti-martingale, never martingale

When the streak hits, the amateur instinct is to increase size to recover faster. This is the martingale, and it is a mathematical guarantee of eventual ruin dressed up as a comeback plan. Doubling into losers means one long-enough streak — the streak that will come — wipes you out entirely.

The professional does the opposite. If size flexes at all, it flexes down in drawdown and up only as equity recovers. That’s anti-martingale, and it’s what fixed fractional gives you for free. The goal isn’t to make the streak back fast. It’s to still be seated when your edge resumes printing.

How big a fraction? Only your numbers know

Here’s the honest limit of any advice, including this. The “right” risk fraction isn’t a universal number someone can hand you. It depends on your system’s real win rate, your average R-multiple, and the worst losing streak your edge actually produces — none of which you can know from a course or a gut feeling. Two traders with different systems will have genuinely different correct answers, and the only credible source is a real sample of your own closed trades.

Estimate your true expectancy from that sample, look at the longest losing run it contains, and size so that run — plus a margin, because your worst streak is almost always ahead of you, not behind — stays comfortably inside your account’s drawdown limits. That’s not a formula you download. It’s a measurement you take.

See it in Shibiki

Shibiki auto-journals every trade, so your win rate, R-multiple distribution, and actual longest losing streak are computed from real fills instead of hopeful memory. In Shibiki, you’d see an edge-health panel per setup with a Wilson confidence interval on the win rate — so a lucky early run doesn’t trick you into sizing as if your edge is bigger than it is. You’d see your R-multiple distribution and the longest observed streak of red, side by side, giving you the raw inputs to pick a risk fraction that survives it. No invented statistics — just the streaks and multiples your own account has already lived through.

A survivor’s checklist

Same small fraction, every trade. Set it, size to it, don’t negotiate mid-session.
Never up-size to recover. The urge to make it back fast is the urge that ends accounts.
Know your worst streak, then assume worse. Size for a run longer than any you’ve seen.
Respect the recovery math. Shallow drawdowns are recoverable; deep ones often aren’t, in time.
Confirm your firm’s drawdown rules before you calibrate anything — static and trailing limits demand different sizing headroom.

The traders who are still funded a year from now aren’t the ones who found a system that never loses. There’s no such thing. They’re the ones who sized so the losing streaks — the guaranteed, unavoidable, perfectly normal losing streaks — were something to endure, not something to fear.