Beginners chase win rate because it feels good to be right. This experiment shows, with the same strategy dialled to two settings, why win rate alone tells you almost nothing about whether you'll make money.
We took one strategy and varied only the reward-to-risk target. Profile A: tight 0.3 R:R targets → a gaudy 77% win rate. Profile B: 2.0 R:R targets → a 'worse-looking' 52% win rate.
Profile A (77% win) came out around break-even — a 1.0 profit factor. Profile B (52% win) produced a 1.22 profit factor and actually grew the account. The lower win rate made more money.
At 0.3 R:R you risk 3 to make 1 — a handful of losers erases dozens of winners, and spread eats the thin margin. At 2.0 R:R each winner pays for two losers, so you only need to be right about a third of the time to profit. Win rate is a vanity metric; expectancy (win rate × average win − loss rate × average loss) is the truth.
Any guru showing you a 90% win rate is almost certainly running tight targets that break even or lose after costs. Ask for the profit factor and the reward-to-risk, not the win rate.