Who this is for — Traders who want to know whether their model earns on average, not only in fortunate periods. It is the base metric for deciding whether to increase size, suspend, or refine a strategy.
In plain terms — Positive expectancy means that each trade, on average, adds value to capital even when many individual trades close at a loss.
Prerequisites — Complete first silver-path (min.: setup, expectancy, win-rate, sample-size). Foundation: bronze-path.
How expectancy is built
Expectancy combines win frequency with the ratio of average gains to average losses. A high win rate with weak payoff can be worse than a low win rate with robust payoff. For this reason read it with edge, not as an isolated number. In practice include real execution costs, otherwise the estimate is optimistic.
Example — System A: 45% win rate, average win 2R, average loss 1R. Even with fewer winning trades, expectancy stays positive. System B: 70% win rate but average win 0.5R and average loss 1.2R — negative expectancy.
Operational reading and limits
Positive expectancy does not eliminate drawdown: it describes the long run, not the next trade. Verify it on updated samples and segment by regime to avoid false confidence. If it stays below zero for weeks, reduce size and reassess application conditions. The link to return-distribution helps understand whether results are stable or concentrated in a few outliers.
Card
- What it is: average expected value per trade.
- How to use it: quality filter before scaling capital.
- Typical mistake: ignoring commissions and slippage in the calculation.
Gold path — Module: Edge and statistical advantage. Part of gold-path.