For a long time, I judged strategies mostly by **win rate**.
Higher win rate = better strategy.
Or so I thought.
After digging into a bunch of backtests recently, what kept showing up was the opposite:
* Some of the *best* performers had **<45% win rate**
* Their edge came from **asymmetric payoffs**, not accuracy
* One or two bad trades usually explained most of the drawdown
* Strategies with “beautiful” stats often collapsed out of sample
What changed my thinking was forcing myself to look at:
* Full return distributions
* Drawdowns vs volatility regimes
* How fragile results were to tiny parameter changes
Once I did that, win rate became almost a **secondary metric**.
Now I treat it more like a personality trait of a strategy, not a quality score.
Curious how others here think about this:
What metric made you *stop trusting* win rate?