Free tool · no email required
Breakeven win rate calculator
Drag the sliders to match your backtest. This tells you the win rate your payoff actually requires, your expectancy after costs, and whether your sample is even large enough for the result to mean anything.
Your strategy
Target distance divided by stop distance.
The percentage your backtest actually produced.
Spread and commission, expressed as a fraction of what you risk.
Sample size of the backtest.
Breakeven win rate
35.0%
What this payoff requires
Your win rate
40.0%
+5.0 pts vs breakeven
Expectancy
+0.150 R
Per trade, after costs
Total over sample
+30.0 R
Across 200 trades
95% confidence interval on expectancy
-0.054 to +0.354 R
Standard deviation per trade is 1.47R. If this interval straddles zero, the backtest has not distinguished your strategy from a coin flip, however good the total looks.
Positive, but the sample can't prove it
Expectancy is positive, but with 200 trades the 95% interval still contains zero — this result is consistent with having no edge at all. You would need roughly 369 trades at this expectancy before that interval clears zero.
How to read the output
Why the breakeven rate matters more than the win rate
A 35% win rate is excellent at 3:1 and ruinous at 1:1. On its own the number carries no information at all. Put it next to the rate your payoff requires and it immediately tells you whether you have an edge or a story.
Why costs move the goalposts
Spread and commission are charged on winners and losers alike, so they raise the required win rate on both sides of the trade. At 2:1 with no costs you need 33.3%. Add a 0.05R round-turn and you need 35.0%. Small numbers decide close cases.
Why sample size decides whether any of it means anything
Trade outcomes are noisy. With 50 trades, a strategy with genuinely zero edge routinely produces a run that looks like a clear winner. The confidence interval is what separates “this worked” from “this is not distinguishable from chance”.
What this calculator cannot tell you
Nothing here checks whether the win rate you typed in is real. If it came from a backtest with lookahead bias, unrealistic fills, or parameters tuned on the same data used to score them, the arithmetic will faithfully compute a confident answer from a corrupt input. That check is the actual work.
The arithmetic, in full
win return = R − c
loss return = −(1 + c)
breakeven p* = (1 + c) / (R + 1)
expectancy E = p(R − c) − (1 − p)(1 + c)
σ² = p(win − E)² + (1 − p)(loss − E)²
95% interval = E ± 1.96 · σ / √n
R is the reward-to-risk ratio, c the round-turn cost expressed in R, p the win rate, and n the number of trades. No hidden adjustments — you can reproduce every number on this page in a spreadsheet.