Guide · Expectancy
The short answer
Expectancy is the average outcome per trade. It is the only number that says whether a strategy earns anything.
Expectancy is the average result of one trade, taken over many. It is the only single number that answers whether a strategy makes money, because it contains both how often you win and how much you win when you do.
Positive expectancy means the strategy earns, on average, per trade. Negative means it loses, however good any individual trade felt.
A strategy wins 40% of the time. Winners average $300; losers average $150.
Thirty dollars per trade, and it loses six times out of ten. Over 200 trades that is $6,000, arrived at through 120 losses. The losing feels like failure the whole way.
Reverse the numbers. A strategy that wins 70% of the time, with $100 winners and $400 losers:
Seven wins in ten, and it bleeds. This is the shape of most strategies that feel good to trade: frequent small wins, occasional large losses, and a comfortable equity curve right up until it is not.
Worth knowing
The two examples differ by $80 per trade and the losing one has a win rate 30 points higher. Any metric that ranks them the other way round — win rate, longest winning streak, percentage of profitable months — is measuring comfort, not edge.
Dollar expectancy depends on position size, so it cannot compare two strategies traded at different sizes. Dividing by the amount risked gives expectancy in R — multiples of the risk taken — which is size-independent:
A strategy with E = +0.2R earns a fifth of what it risks, per trade, whatever the account. That is the number worth tracking, and the one to record in a trading journal rather than recomputing from memory.
Expectancy is an average, and averages say nothing about order. A strategy with +0.2R expectancy can still deliver twelve losses in a row, and whether you survive that depends on position size, not on expectancy.
It also assumes your sample is representative. Twenty trades tell you almost nothing; the standard error on a win rate from twenty observations is wide enough to contain both of the examples above. See how drawdown affects recovery for the other half of the problem.
Profit factor is gross profit divided by gross loss. It is the other statistic quoted alongside expectancy, and the two disagree in a way worth understanding.
A strategy with profit factor 1.5 wins 50% more than it loses in total, but that says nothing about how many trades produced it. Twenty trades at profit factor 1.5 and two thousand at profit factor 1.5 describe very different confidence in the same number.
Expectancy in R has the same weakness and one advantage: it is per trade, so multiplying by a trade count gives a projection directly, and a projection makes the sample size impossible to ignore.
Expectancy inherits its uncertainty from the win rate and the two averages, and all three are estimated. Early on it moves a lot, and the movement is the estimate rather than the strategy.
| Trades so far | Effect of one 5R winner |
|---|---|
| 20 | Moves expectancy by about 0.25R |
| 100 | About 0.05R |
| 500 | About 0.01R |
At twenty trades a single good outcome swamps everything else, which is why a strong first month says very little. The figure is worth acting on somewhere past a few hundred trades, and worth recording from the first one.
A strategy with +0.2R expectancy over five hundred trades still produces losing months, and the frequency of them surprises people who have only seen the average.
Neither is evidence of a problem. The average is a long-run property, and the path to it is noisy enough that any single month carries almost no information about whether the edge is intact.
The average result of one trade taken over many: E = (win rate × average win) − (loss rate × average loss). It is the only single number that says whether a strategy earns anything, because it contains both how often you win and how much.
Yes. A 40% win rate with $300 winners and $150 losers earns $30 per trade. A 70% win rate with $100 winners and $400 losers loses $50 per trade. Win rate on its own determines nothing.