Risk management

Expectancy Calculator: Does Your Strategy Have an Edge

Astron Team · 09/24/2026 · 4 min read

A strategy can win most of its trades and still slowly drain the account. The opposite is also true: it's possible to lose more than half of the entries and still end the month in profit. Win rate alone doesn't answer whether a strategy has a real edge. What answers that question is mathematical expectancy, also called expectancy.

Expectancy shows, on average, how much a strategy gains or loses per trade, based on the history of results. A positive number is evidence of a statistical edge within the observed sample, not a guarantee of future profit.

The expectancy formula

The best-known version of the formula is:

Expectancy = (win rate x average gain) - (loss rate x average loss)

This formula works well when every trade ends in a gain or a loss, and when the averages already include costs like commissions. To make the calculation even clearer, you can also calculate net expectancy directly from the totals:

Net expectancy = [(number of wins x average gain) - (number of losses x average loss) - total costs] / total trades

A worked example in reais

Suppose a trader reviewed 40 trades from the same setup, all with the same planned risk per entry, and arrived at the following numbers:

• Winning trades: 18
• Losing trades: 22
• Average gain per winning trade: R$ 150.00
• Average loss per losing trade: R$ 80.00
• Average planned risk per trade: R$ 100.00

The calculation starts by separating total gains and losses:

• Total gain: 18 x R$ 150.00 = R$ 2,700.00
• Total loss: 22 x R$ 80.00 = R$ 1,760.00
• Net result of the sample: R$ 2,700.00 - R$ 1,760.00 = R$ 940.00
• Expectancy per trade: R$ 940.00 / 40 = R$ 23.50

Expressing this result in multiples of risk (R), using the average risk of R$ 100.00 per trade, the expectancy comes out to +0.235R per trade. This means that, on average across this sample, each trade returned about 23.5% of the planned risk. The profit factor for this same sample, calculated as total gain divided by total loss, comes to 2,700 / 1,760, or approximately 1.53.

A 45% win rate can still be positive

Notice that the win rate in the example above is just 45% (18 out of 40 trades), below half. Even so, the expectancy is positive, because the average gain is almost twice the average loss. This is why asking only what a strategy's win rate is tends to lead to wrong conclusions about its quality.

Don't forget about costs

Commissions, broker fees, and possible costs of holding a position open overnight reduce net expectancy. A strategy with a gross expectancy of R$ 6.00 per trade can turn negative after deducting R$ 7.00 in average costs per trade. Whenever possible, use the real costs charged by the broker, not an optimistic estimate, so you don't artificially inflate the result.

A second example, this time with negative expectancy

To make the contrast clear, look at a second case with 50 trades: 21 winners with an average gain of R$ 90.00 and 29 losers with an average loss of R$ 75.00. Total gain comes to 21 x R$ 90.00 = R$ 1,890.00, and total loss to 29 x R$ 75.00 = R$ 2,175.00. The net result of the sample is negative, at R$ 285.00, which gives an expectancy of R$ 1,890.00 minus R$ 2,175.00, divided by 50 trades, that is, an average loss of R$ 5.70 per trade. Even with a 42% win rate, close to the range considered normal for many strategies, the relationship between average gain and average loss wasn't enough to sustain a positive result.

How to use this day to day

How many trades are enough to trust the number

There's no universal magic number. A sample below 30 trades is usually considered too small for any solid conclusion. Between 30 and 100 trades, the result already gives a first estimate, but it can still change quite a bit with a few extreme results. Above 100 or 200 trades, the number starts to become more informative, even though it's never definitive proof that the edge will keep existing in the future.

What to do after calculating expectancy

If the expectancy comes out negative after counting real costs, it doesn't make sense to discuss increasing position size before fixing the strategy. If it comes out positive, it's worth testing how robust the result is: remove the single largest winning trade from the sample and recalculate. If the expectancy turns negative with just that removal, the result may be relying too heavily on one outlier trade, and deserves more caution before increasing position sizes.