I used to look at an 80% win rate and think, “Well, that one is obviously good.” It is not that simple. If eight wins make a dollar each and two losses cost ten dollars each, I still lost money—and that is before fees. A strategy can win only four out of ten and still make money if those four wins are big enough.
The number I actually care about
Traders call it expectancy. It is just a way to combine how often I win with how much I usually win or lose:
Expectancy = (win rate × average win) − (loss rate × average loss)
Then I still have to subtract fees, spread, slippage, and slow fills. A tiny paper profit can disappear fast once the trade has to happen in the real world.
What has to sit next to win rate
- Net expectancy. What each trade makes or loses on average after costs.
- Average win and average loss. Whether the wins are actually big enough to pay for the losses.
- Payoff ratio and profit factor. Two more ways to compare the dollars won with the dollars lost.
- Maximum drawdown. The biggest hole the bankroll fell into along the way.
- Trade count and cost drag. Whether there is enough evidence and how much the exchange ate.
Watch out for the one giant ass-kicking
An average can hide a disaster. Some strategies stack up little wins and then get flattened by one ugly move. So I also want to see the worst losses, how far trades moved against us, whether too much money sat in one place, and what happened during market crashes.
Five lucky trades do not prove shit
Great numbers from a handful of trades might just be luck. A slow strategy may need more time. A fast one may fire twenty trades that are really the same market move wearing different clothes.
Win rate belongs on the dashboard. It just does not get to tell the whole story by itself.