In a trading journal, estimated expectancy is the average result per trade in the recorded sample. A positive historical average is useful evidence to investigate; it is not a promise that the next trades will have the same average.

Calculate It From Net Results

The simplest calculation is:

Sample expectancy = total net trading P&L / number of trades

Alternatively, when every trade is classified as a win or loss:

Expectancy = (win fraction × average win) − (loss fraction × average loss)

Use a positive magnitude for average loss. If there are breakeven trades, include them in the total count; the loss fraction is then not automatically one minus the win fraction. Costs can turn a gross breakeven trade into a net loss.

For a hypothetical sample with 45% winners averaging $200 and 55% losers averaging $100, the gross estimate is:

0.45 × $200 − 0.55 × $100 = $35 per trade.

If average trading costs are $8 per trade and were not already included, the net estimate is $27. Multiplying that number by a hoped-for weekly trade count would add assumptions about opportunity frequency and unchanged performance; it is not an income forecast.

Use Realised Results, Not Planned Targets

A planned 1: 2 reward-to-risk setup does not imply an average win twice the average loss. Partial exits, skipped targets, gaps and discretionary decisions change realised results. Use the actual P&L from the broker record. Those fills already reflect execution prices: do not deduct estimated slippage a second time from realised P&L.

For a historical test, model commissions, fees and plausible execution differences because ideal chart prices are not fills. Keep simulated results distinct from live records.

To compare trades taken at different sizes, also calculate each outcome in R, dividing net P&L by that trade's initial planned dollar risk. Average those individual R results. Dividing total P&L by one convenient risk amount can misstate a sample whose sizing changed.

Why Win Rate Alone Is Insufficient

A hypothetical 70% win rate with $50 average wins and $200 average losses produces −$25 per trade before costs. A 30% win rate with $500 average wins and $100 average losses produces +$80 before costs.

Neither example is a recommended strategy. They show why win frequency and payoff size must be assessed together. See basic trading statistics for the related breakeven calculation.

Assess the Estimate Before Acting on It

Keep the strategy version, sample dates and trade count beside the estimate. Inspect whether a few unusually large wins dominate it, whether losses cluster and whether the result changes across market conditions. A fixed count such as 30,100 or 500 trades cannot by itself establish statistical confidence.

Use a separate period that was not used to select entries or optimise targets. If you revise the strategy after seeing the results, evaluate that new version separately. Increasing a target or tightening a stop does not automatically improve net expectancy because it can change the frequency and size of both wins and losses.

Frequently Asked Questions

What is a good expectancy? There is no universal R benchmark. A positive sample estimate needs context: uncertainty, costs, drawdown, available opportunities and the risk of losses larger than the initial stop. Compare estimates only when trade grouping and cost treatment match.

Can a positive estimate coexist with a losing month? Yes. An average does not describe the sequence or spread of outcomes. Also, a historical estimate may cease to describe current conditions.

How often should I calculate it? Use a regular review interval that fits the strategy's trade frequency, and investigate execution or risk-limit problems immediately. Avoid reacting to every small fluctuation or ignoring a serious problem until a target sample count is reached.

Risk Disclaimer: Trading involves substantial risk of loss. Past performance is not indicative of future results. See our full risk disclaimer.