Win Rate Confidence Interval Calculator
Calculate an observed positive-outcome rate and a two-sided Wilson score interval from entered wins, losses and breakevens. The interval describes the supplied sample under the stated model; it does not verify a trading edge or predict future results.
Enter observed outcomes
Counts must be whole numbers. Breakevens remain in the sample denominator.
Observed rate = wins / (wins + losses + breakevens). Total observations cannot exceed 10,000.
Observed interval
Observed Win-Rate Interval 1.0.0.
Out of date — recalculate after changing an input.
How the Wilson interval is calculated
The observed rate is the entered win count divided by all entered outcomes. The two-sided Wilson score method then adjusts the center and width using the selected confidence level and sample size.
Wilson denominator = 1 + z² / n
Limits = adjusted center ± adjusted half-width
The calculator uses governed z-values for 90%, 95% and 99% intervals, keeps full precision during calculation and applies no continuity correction.
Worked example from the audited fixture
Enter 40 wins, 50 losses and 10 breakevens with a 95% confidence level. Breakevens remain in the denominator, so the observed positive-outcome rate is 40 ÷ 100 = 40%. Using the governed 95% z-value, the two-sided Wilson interval is approximately 30.9401% to 49.7997%, a width of 18.8596 percentage points.
How to interpret the result
The interval displays sampling uncertainty under the entered classifications and binomial model. It is not a 95% probability that the next trade wins or proof that the underlying strategy has an edge. Missing trades, dependence, regime changes and inconsistent breakeven treatment can make the statistical calculation a poor description of future conditions.
Assumptions and limits
- The tool does not verify whether trades were sampled randomly or recorded completely.
- Independence, stationarity and consistent win/loss/breakeven classification are not established.
- Selection bias, changing market conditions, trading costs and execution are outside the interval.
- A confidence interval is not a probability that a future trade will win.
- No result is a strategy grade, verified edge, performance forecast or trading recommendation.
Frequently asked questions
- It is the number of entered wins divided by all entered wins, losses and breakevens.
- A breakeven is an observed non-win in this contract, so it remains in the denominator rather than being removed.
- It is a binomial-proportion interval that adjusts its center and width for sample size and the selected confidence level.
- The tool supports 90%, 95% and 99%. A higher selected level generally produces a wider interval; the page does not recommend one.
- No. The result does not verify data quality, independence, stationarity, costs or future persistence.
- Yes, but the interval can be wide. The page reports that uncertainty rather than hiding the result.
- Breakeven win rate is a payoff-and-cost threshold. This route instead summarizes entered observed counts.
- No. It contains no next-trade forecast, signal or recommendation.
Sources and methodology
- NIST/SEMATECH — Confidence Limits for a Proportion — Wilson score interval formula and interpretation context.
- CFTC — Trading System Claims Advisory — hypothetical and past-performance limitations.
Continue the evidence workflow
Verify account and execution records
Before classifying outcomes, confirm the relevant broker entity, account currency, costs and statement basis.
Risk and affiliate disclosure: Leveraged forex and CFD trading can result in substantial losses. These are affiliate links, so ForexMT4Indicators.com may receive compensation if you register or trade through them, at no additional cost to you. Availability and terms vary by jurisdiction and broker entity.