Who this is for — Traders watching only win rate or monthly P&L. Expectancy tells if the **system** earns per trade on average.
Expectancy is average expected value of a trade in the sample: combines win frequency and average win vs loss size. R units preferred to compare setups and periods.
In plain terms — How much you «print» on average each button press — win or loss — over 100+ trades.
Formula
EV = (Win% × Avg Win) − (Loss% × Avg Loss)
In R (recommended):
EV = (Win% × Avg R win) − (Loss% × Avg R loss)
| Input | Example |
|---|---|
| Win 40%, avg +2R | 0.40 × 2 = 0.8R |
| Loss 60%, avg −1R | 0.60 × 1 = 0.6R |
| EV | +0.2R/trade |
100 trades ≈ +20R expected (before variance).
Operational verdict
- EV < 0 on adequate sample size → stop live, fix system
- EV > 0 → execution discipline is bottleneck, not «find better setup» daily
Typical mistake — EV on 15 trades or bull period only — illusory.
Example — 60 trades: 38% win, avg +2.3R / −1R → EV ≈ +0.25R. Goal: keep execution, not raise win rate at all costs.
Summary sheet
- Unit: R per trade.
- Min sample: 50–100+ trades.
- Pair: Win rate + payoff.
Silver path — Metrics module. Index: Silver path.