By Cree Lawson
Nine winning trades out of ten sounds like a comfortable way to trade. I would still want to see the tenth trade before deciding what that record means.
A win rate counts how often trades finish ahead. It doesn't tell us how much the winners make or how much the losers give back. Both sides matter when you're deciding whether an Expert Advisor, or EA, deserves further testing.
Consider an illustrative set of 100 trades. Ninety make $10 each. Ten lose $100 each. The winning trades contribute $900, the losses total $1,000, and the result is a $100 loss before any additional costs. The win rate is 90% throughout.
Now consider a different illustrative set. Forty trades make $30 each and sixty lose $10 each. That produces $1,200 in gains and $600 in losses, for a $600 gain before additional costs. Its win rate is 40%.
Neither example is a recommendation or an actual EA result. They show why I want the win rate, average win and average loss on the same page.
Read the losing side of the report
For a historical sample with no breakeven trades, the average result per trade can be calculated as the winning fraction times the average win, minus the losing fraction times the average loss magnitude. Use consistent treatment of trading costs so you don't omit them or subtract them twice.
That calculation describes the sample. It doesn't establish what the next trade will earn.
I would then look beyond the averages. Was the largest loss much bigger than the others? Did losses arrive close together? How long were positions open before they closed? Averages can conceal a result that depends heavily on a few unusual events.
Find out what counts as one trade
An EA may open several related positions and close them as a basket. Partial closures can also make a report harder to interpret. Before comparing two win rates, establish how each report counts trades and whether the observations represent separate decisions.
I would inspect the equity curve as well. A completed winner can have spent time carrying a substantial floating loss. Its final outcome doesn't tell you what the account had to support while it was open.
Keep the number in its place
A high win rate can be a legitimate feature of a strategy. It deserves a clear explanation of the size and timing of losses, just as a lower win rate deserves an explanation of how its winners compensate for them.
For your next EA review, save the win rate, average win, average loss, largest loss and equity drawdown together. Then look at the underlying trades that produced those figures. The MT5 testing-report reference defines the reported fields.
When I see a 90% win rate, the next thing I want to understand is what happens during the other 10%.
Educational discussion. Examples are hypothetical, and historical results do not establish future performance. Trading can result in substantial loss.
