Quant
Expected value vs win rate — why a 40% strategy wins
Ask someone how good their trading strategy is and they'll usually quote a win rate: "I'm right 70% of the time." It's the most intuitive number and one of the most misleading. A strategy that wins 70% of the time can lose money, and one that wins 40% of the time can print. What actually matters is expected value — and it depends just as much on the size of wins and losses as on how often they happen.
The number that pays: expected value
Expected value (EV) is the average outcome per trade if you repeated it forever. It weighs each outcome by both its probability and its size:
EV = (win% × average win) − (loss% × average loss)
Take a strategy that wins only 40% of the time, but wins are large and losses are cut short — say +30 points when right, −15 when wrong:
Wins contribute 0.40 × 30 = +12; losses cost 0.60 × 15 = −9. Net EV is +3 points per trade. You're wrong more often than right and still make money, because you win big and lose small.
Why win rate seduces
A high win rate feels good — being right is pleasant, and a string of small wins is comforting. But it hides the tail. Strategies that win 95% of the time often do so by risking a lot to make a little (selling insurance, picking up pennies) — and the rare 5% loss can wipe out months of gains. High win rate, deeply negative EV. The comfort is exactly what makes it dangerous.
Winning often ≠ making money. What matters is how much you win when right versus lose when wrong. Big wins and small losses beat a high win rate with the sizes reversed.
What to actually track
- Expected value per trade — is it positive after costs? Everything starts here.
- Payoff ratio — average win ÷ average loss. A high ratio lets a low win rate thrive.
- Both together — win rate is only meaningful alongside the payoff ratio. Neither means much alone.
A positive EV is the price of entry, not a guarantee — you still have to survive the variance to collect it, which is where position sizing and drawdown come in, and why a Sharpe ratio (which rewards smooth EV) matters more than raw win rate. Chase expected value, size it so the losing streaks can't ruin you, and let the win rate be whatever it needs to be.