“I win 70% of my trades.” It’s the most quoted stat in trading and the most useless one, because it deliberately hides the other half of the story: what those wins are worth against what the losses cost.
Why win rate on its own is meaningless
Win rate is the percentage of your trades that close green. That’s all it measures — frequency, not profit. It says nothing about the size of the wins versus the losses, which is where money is actually made or lost.
You can win 90% of the time and still go broke: win ninety small trades for +0.2R each (+18R) and lose ten trades for −2R each (−20R), and you’re net −2R despite a spectacular-looking hit rate. Flip it — win 30% at +4R and lose 70% at −1R — and you’re comfortably profitable while “losing most of the time.”
The lesson: a win rate quoted without its payoff is a number designed to impress, not to inform. Anyone who leads with win rate and goes quiet on average win vs average loss is telling you the incomplete half on purpose.
The seesaw: win rate vs payoff size
Win rate and risk-reward (the size of your average win relative to your average loss, measured in R) sit on opposite ends of a seesaw. Push one up and the market usually pushes the other down.
- Want a higher win rate? Take profit sooner and give trades more room to breathe. You’ll win more often — but each winner is smaller, dragging your R:R down.
- Want a bigger R:R? Hold for distant targets and cut losers fast. Your winners get fat — but you’ll be stopped out more often reaching for them, dragging your win rate down.
Almost no strategy escapes this. A method with both a high win rate and a high R:R is either a very rare genuine edge or, far more often, a small sample that hasn’t met a bad month yet. When someone claims both, the responsible question is “over how many trades?” — the tradeoff reasserts itself as the sample grows.
The breakeven line that ties them together
The two numbers aren’t independent — they’re joined by a hard mathematical floor called the breakeven win rate: the minimum hit rate a given risk-reward needs just to avoid losing money.
Breakeven win rate = 1 ÷ (1 + reward:risk)
- At 1R, you need to win more than 50%.
- At 2R, you need more than ~33%.
- At 3R, you need more than 25%.
Your real question is never “is my win rate good?” It’s “does my win rate clear the breakeven line for the R:R I actually trade?” A 45% win rate is fantastic at 2R and a disaster at 0.5R — the same win rate, opposite verdicts, decided entirely by payoff. The risk-reward calculator gives you the breakeven line for any ratio, so you can check your hit rate against the bar it actually has to clear.
High-win-rate vs high-R:R styles compared
Neither end of the seesaw is “correct” — they’re different jobs with different demands. What matters is that your temperament matches the style you pick.
| High win rate | High risk-reward | |
|---|---|---|
| Typical win rate | 60–80% | 30–45% |
| Typical R:R | Below 1.5R | 2.5R and up |
| Feels like | Frequent small wins | Long droughts, occasional big wins |
| Main danger | One oversized loss erasing many wins | Cutting winners early; skipping trades in a drought |
| Emotional demand | Discipline to keep losers small | Patience to sit through losing streaks |
| Common styles | Mean-reversion, scalping | Trend-following, breakouts |
The failure mode is always the opposite discipline. High-win-rate traders die by letting one loss run because they’re used to being right. High-R:R traders die by snatching a +1R when the plan said +4R, because a string of losers wore down their patience. Know which mistake your style tempts you into, and build the guardrail there. Journals like Edgewonk exist precisely to surface these patterns after the fact — the same job Shibiki’s auto-journal does automatically as trades close, so you don’t rely on remembering to log the one loss you’d rather forget.
Judge the pair together through expectancy
Stop evaluating win rate and R:R as separate scores. There’s one number that fuses them into a verdict: expectancy — your average result per trade in R.
Expectancy = (Win% × Avg win in R) − (Loss% × Avg loss in R)
Run both example styles through it:
- High win rate: 70% × 1.0R − 30% × 1.2R = 0.70 − 0.36 = +0.34R
- High R:R: 35% × 3.0R − 65% × 1.0R = 1.05 − 0.65 = +0.40R
Two completely different-looking systems, roughly the same real edge. Expectancy is the only fair judge because it holds win rate and payoff in the same frame — raise either lever and expectancy tells you whether it was actually worth it. The expectancy calculator and the deeper expectancy explainer walk through the full math.
The catch is trust. A tiny sample can flatter either style — one lucky runner spikes your R:R, one hot week inflates your win rate. That’s why Shibiki reports live edge health with a Wilson confidence interval rather than a single expectancy headline: it shows the range your true edge is likely to sit in, tightening as the sample grows, so you can tell a real, durable edge from a small-sample mirage before you scale it up.
Related: Trading expectancy · Expectancy calculator · Risk-reward calculator