Trading Expectancy Explained: Measuring an Edge Over Many Trades

Expectancy should be reviewed alongside the distribution of results. Two strategies can have the same average expectancy while one has many small outcomes and the other has rare large outcomes. The second may require more cash capacity and emotional tolerance because losing streaks can be longer. Group trades by setup, market condition, and time so a blended average does not hide a weak segment. Compare the result before and after costs, then check whether the sample contains enough observations for the summary to be meaningful. If a strategy changes, start a new record rather than combining old and new rules. Position size should be selected from the loss that can be accepted, not from the average profit that is hoped for. This keeps expectancy in its proper role as a measurement of a process rather than a target that encourages excessive activity.

Trading expectancy is the average amount a strategy may gain or lose per trade under a defined set of results and costs. It combines the chance of a win with the size of an average win and combines the chance of a loss with the size of an average loss. A positive sample expectancy is not proof of a durable edge. Results vary, the sample may be too small, and market conditions can change. Expectancy is useful because it shifts attention from win rate alone to the full payoff distribution. A strategy with many small wins and occasional large losses can have negative expectancy, while a strategy with fewer wins can be positive if its winners are sufficiently larger. Costs and execution belong in the calculation.

Expectancy formula

Expectancy per trade = (win probability x average win) - (loss probability x average loss) - average cost. If a strategy wins p of trades and loses 1 - p, use the average net win and net loss after commissions, spread, and financing where applicable. Break even win rate before costs is average loss divided by average win plus average loss. With a $100 average win and $80 average loss, break even is $80/$180 = 44.4444%. A win rate above that can still lose money after costs. Expectancy is measured in dollars, account percentage, or units of risk. Use one unit consistently. Position sizing changes dollar expectancy while the underlying per risk unit expectancy can remain the same.

Worked 100 trade sample

Suppose a 100 trade sample has 45 wins averaging $120 gross and 55 losses averaging $80 gross. Gross expectancy is 0.45 x $120 - 0.55 x $80 = $54 - $44 = $10 per trade. If average spread, commission, and financing cost is $3 per trade, net expectancy is $7. Total net result for the sample is $700 before any unusual execution difference. The gross break even win rate is $80/($120 + $80) = 40%, so the observed 45% win rate is above that gross threshold. A loss of $100 on the next trade does not invalidate the sample expectancy, but it changes the account path. The result is an average, not a scheduled payment.

Variance and drawdown

Expectancy does not describe the order of trades. Five losses in a row can occur in a strategy with positive expectancy. If each loss is 1% of current equity, five consecutive losses leave 0.99^5 = 95.099% of starting equity, a decline of about 4.901%, before other costs. If risk is a fixed dollar amount, the percentage loss changes as equity changes. If risk is a fixed percentage, the dollar loss changes with equity. Win rate and average payoff can also vary by market regime. Track a distribution of outcomes, maximum drawdown, losing streaks, and results by setup, session, and condition. A positive average does not establish how much capital can be lost before the edge appears.

Sample quality and assumptions

A result from ten trades has little statistical information compared with a well documented larger sample, and even a large sample can be biased. Include losses omitted by discretionary exits, inactive periods, spread changes, slippage, rejected orders, financing, and data errors. Avoid changing rules after seeing each result unless the new rule begins a new test. Separate development data from evaluation data where possible. Market conditions can make a strategy’s observed edge temporary. The formula assumes a stable distribution and independent enough observations for the summary to be meaningful, but trades can be correlated. Do not use a high backtest expectancy to promise future returns or to justify unlimited leverage.

Practical use

Keep a trade journal with planned risk, realised result, costs, setup, and exit reason. Calculate gross and net expectancy separately. The forex compound calculator can show how a selected average return might compound, but it should not replace a distribution of wins and losses. The forex position sizing guide explains how risk per trade connects to account size. Use the debt payoff calculator only for debt decisions, not as a trading forecast. Set a maximum account drawdown, reduce activity when data shows a changed regime, and review costs. Treat the expectancy estimate as a measurement to update, not a target that must be forced by taking more trades.

Common mistakes

• Using win rate alone without average win, average loss, and costs.
• Calling gross expectancy positive after omitting spread, commission, or financing.
• Assuming the next sequence will resemble the average sequence.
• Estimating expectancy from too few or selectively reported trades.
• Changing position size to recover from a losing streak.
• Treating backtested expectancy as a guarantee of future performance.

Conclusion

Trading expectancy measures the average net result implied by win probability, average win, average loss, and costs. In the 100 trade example, $10 gross expectancy became $7 after a $3 average cost. The strategy could still experience losing streaks and drawdown, and the sample might not represent future conditions. Track complete results, measure variance and correlations, size positions from an affordable loss, and update the estimate when evidence changes. Compounding a positive average is a scenario, not a promise. Expectancy is most useful as one risk and process metric inside a broader trading plan that limits loss and respects uncertainty.