Nothing feels more justified than pressing size after five green trades in a row. Nothing empties a funded account faster than being wrong about why those five were green.
The streak doesn’t change the next trade
Start with the math, because your emotions won’t. If your setup wins some fraction of the time, a run of wins does not raise the probability that the next one wins. Independent trades don’t remember the streak. The market has no idea you’re up five.
This is the gambler’s fallacy wearing a trader’s hat. A win streak feels like accumulating evidence that “it’s working right now,” and sometimes conditions genuinely are favorable. But most streaks are ordinary variance — the normal clustering that any positive-expectancy system produces. Flip a slightly-weighted coin long enough and you’ll get runs. Runs are not a signal to lever up.
The danger is that sizing up on a streak stacks your largest bets at a random moment. When the inevitable regression to the mean arrives, it lands on your biggest positions — turning a normal cooldown into a drawdown that undoes the whole run and then some.
Anti-martingale isn’t wrong — it’s just conditional
To be fair, there’s a legitimate case for increasing size in strength. Anti-martingale — risking more when you’re winning, less when you’re losing — is a real, defensible approach. If your edge is genuinely stronger in certain regimes, and your winning trades cluster because conditions are actually favorable, then leaning in captures more of a real edge.
The problem is never the concept. It’s that traders apply it on feeling rather than evidence:
- Sizing up because they’re confident, not because the regime measurably favors the setup.
- Adding risk right as the streak is statistically most likely to end.
- Scaling the bet faster than any real edge could possibly justify.
The honest position: increasing size after wins can be correct or catastrophic, and the streak alone can’t tell you which. Only your history of what happened after similar streaks can. Model any size change against a position size calculator so the increase is a deliberate step, not an adrenaline-driven guess.
On a funded account, the asymmetry is brutal
Personal capital forgives a streak-fueled mistake — you rebuild. A funded account often doesn’t. Sizing up right before variance turns can drop you through a daily loss limit or into your trailing drawdown in a single session, ending an account you spent weeks and an evaluation fee earning.
The trailing drawdown mechanic makes this especially cruel after a streak: your winning run pushes your equity high up, the drawdown floor ratchets right up behind it, and now your enlarged position is being measured against a tighter buffer than you had before the streak started. You feel richer and more protected. You’re actually more fragile. Never assume a firm’s exact limits from memory — confirm the live numbers with your firm, because a streak is precisely when overconfidence skips the fine print.
Confidence is not the same as edge
Here’s the core confusion. Confidence is a feeling. Edge is a number. They drift apart constantly — never more than during a hot streak, when confidence runs far ahead of anything the data supports.
The question that actually matters isn’t “do I feel good about this?” It’s “does my expectancy improve or degrade when I size up after wins, measured across a real sample?” That’s answerable. It’s just that almost nobody answers it — they let the feeling decide and call it discipline.
Ground it in expectancy and R-multiples: if your average result stays healthy as post-streak size rises, anti-martingale is earning its keep. If your expectancy quietly craters on the sized-up trades, your “confidence” was overexposure with a better story. Run the numbers through an expectancy calculator instead of trusting the glow.
See it in Shibiki
This is exactly the comparison that has to come from data, not gut. In Shibiki, every trade is auto-journaled, so you’d see a side-by-side of your normal-size trades versus your post-streak sized-up trades — each with its own expectancy and a Wilson confidence interval so you can tell whether the difference is a real effect or just too few trades to trust yet. Picture an R-multiple distribution for each group: if the sized-up trades show a fatter left tail — bigger losses without proportionally bigger wins — the app is showing you, in your own numbers, that the streak was talking and you were listening.
A discipline that survives the streak
You don’t have to swear off sizing up. You have to make it a rule instead of a mood:
- Decide size changes in advance, tied to measurable conditions, not to how many greens are behind you.
- Cap the step. Even a justified increase should be a modest, pre-planned notch — never a double-up.
- Never let a streak push you past your risk budget or near a firm limit you haven’t reconfirmed.
- Review afterward. Let the post-streak expectancy tell you whether to keep the behavior or kill it.
The traders who last aren’t the ones who never feel the pull to press winners. They’re the ones who make that decision cold, against their own numbers, before the streak ever starts — so the confidence has to earn its size instead of just borrowing it.
Related: Position size calculator · Expectancy calculator · R-multiple explained