Trading expectancy calculator — gross and net

An 80% win rate with an average win of 0.2R and an average loss of 1R produces −0.04R per trade. A 35% win rate with an average win of 3R and a 1R loss produces +0.40R. Expectancy combines win rate and payoff size; net expectancy also deducts execution costs. A high win rate alone does not establish an edge. The calculator reports gross and net expectancy, both break-even thresholds, a modelled total over 100 trades and an approximate Wilson interval for win rate.

The calculator is loading. If it does not appear, enable JavaScript and reload this page.

Expectancy in R per trade

E gross = p × average win − (1 − p) × average loss; E net = E gross − average cost.

Check the example: 45% × 1.8R − 55% × 1R = +0.260R gross. After 0.10R average cost, expectancy is +0.160R net. Break-even win rate is 1 / 2.8 = 35.7% before costs and 1.1 / 2.8 = 39.3% after costs. The theoretical total over 100 trades is +16R when the monetary size of 1R is constant; this is not a prediction of any particular sequence.

Estimate cost in R by dividing spread, commission and typical adverse slippage expressed in a common monetary or pip unit by the initial risk in the same unit. For example, 1.5 pips of cost divided by a 15-pip stop is 0.10R. Avoid double counting: if average wins and losses already use net journal results, enter zero additional cost. Record positive magnitudes for both average wins and average losses.

How to calculate expectancy manually

Multiply the winning-trade proportion by the average winning R-multiple, then subtract the losing-trade proportion multiplied by the average losing R-multiple. Deduct average execution cost per trade only when it is not already included. Costs are incurred on winning and losing trades alike. Use actual journal data and a consistent definition of initial risk; do not fill the model with hoped-for results. For a separate break-even category, the two-outcome model requires adaptation: either include zero-result trades consistently in the non-winning group’s average, or calculate the weighted mean directly from all trade outcomes.

The uncertainty interval concerns win rate, not expectancy itself. At 60% wins in 30 independent trades, the 95% Wilson interval is approximately 42–75%. An entered rounded win rate can imply a fractional win count, so the interval is an approximation. For precise inference use the actual win count. Average payoff sizes also have uncertainty, and trade clustering, changing market conditions or selection bias can invalidate simple independent-trial assumptions.

Frequently asked questions

What is a good expectancy?

A positive net value is a necessary condition for a profitable model, but it is not evidence by itself that the strategy will remain profitable. Frequency, drawdowns, payoff dispersion and capacity matter. At a constant monetary R, +0.1R over 200 trades has a different expected annual total from +0.5R over 12 trades. The calculator’s 100-trade total is a theoretical comparison at fixed R, not a compound-growth forecast.

Why does sample size matter?

A small sample leaves substantial uncertainty. Thirty trades or a hundred trades are useful review milestones, not certificates of an edge. Keep collecting out-of-sample and forward-test evidence, inspect the distribution of wins and losses, and check that execution assumptions remain realistic. A 95% confidence interval describes the statistical procedure; it is not a 95% guarantee that future win rate lies in the displayed range.

Related tools and course chapters

All calculations run in your browser. Outputs depend on the quality and consistency of inputs; gross results or small samples can mislead. This tool is an educational model, not investment advice.

Statistical reference: NIST: confidence intervals for proportions. Reviewed September 2026.

Model scope and sample interpretation

Average wins and losses are positive magnitudes in R; cost is an average per trade. With 45%, 1.8R, 1R and cost 0.1R, the result is 0.26R gross and 0.16R net. The after-cost break-even rate is 39.29%, and 100 trades at a constant monetary R have an expected total of 16R. Zero-result trades need consistent treatment: include them in the non-winning group’s average or compute the mean directly from all outcomes. The Wilson interval concerns win rate only and assumes independent trials; it does not cover payoff uncertainty or market changes. A rounded win rate can imply fractional wins, making the interval approximate. Neither 30 nor 100 trades guarantees an edge.

NIST: confidence intervals for proportions