Why win rate is the most overrated number in trading
A 90% win rate can bankrupt you. A 40% win rate can make you rich. Here is the arithmetic that decides which — and why the number everyone brags about barely matters.
- Win rate is meaningless without reward-to-risk. Together they define your expectancy — the only number that decides whether you make money.
- A 40% win rate at 2R is more profitable than a 60% win rate at 1R.
- High win rates hide fat-tailed losses — the "one bad trade" that erases fifty good ones.
Ask a room of traders how they're doing and most will answer with a win rate. "I win seventy percent of my trades." It sounds like skill. It sounds like an edge. It is, on its own, almost meaningless — and the belief that it matters is one of the most expensive misconceptions in retail trading.
The number that actually matters
What decides whether you make money is not how often you win, but how much you win when you're right versus how much you lose when you're wrong, weighted by how often each happens. That single blended figure is called expectancy, and every honest trading result reduces to it.
Expressed in R — multiples of the amount you risk per trade — this strips out account size and lets you compare any two strategies directly.
A worked example
Consider two traders. Anaya wins 60% of the time but only makes as much as she risks on each winner (1R). Ben wins just 40% of the time, but each winner returns twice his risk (2R). Intuition says Anaya is the better trader. The arithmetic disagrees.
Anaya: 0.60 × 1R − 0.40 × 1R = +0.20R per trade. Ben: 0.40 × 2R − 0.60 × 1R = +0.30R per trade.
Ben wins far less often, yet earns 50% more per trade. Over a thousand trades that gap compounds into an entirely different account. The lesson is uncomfortable for anyone who has been optimising for the feeling of being right: being right is not the goal. Being paid is.
The hidden danger of high win rates
There's a darker side. Strategies engineered for high win rates — selling options, martingale averaging, grid systems — often do it by taking many small wins and hiding a rare, enormous loss. The equity curve looks flawless right up until the day it doesn't.
This is why we simulate. A win rate tells you nothing about the shape of your losses. Run any strategy through the Strategy Reality Check and watch what the unlucky 10th-percentile path does. If a single bad streak ends the account, the win rate was never the point.
Each faint trail is one simulated future for this strategy. The bold line is the median — half of runs finished above it, half below. The shaded band spans the lucky 10th to unlucky 90th percentile, so an outcome inside the band was reasonable to expect; outside, less so.
What to do with this
Stop reporting your win rate as if it were a grade. Track your average win in R, your average loss in R, and the resulting expectancy. Then protect the assumption that your edge holds by sizing every position so that no realistic losing streak can end you. Win rate is a vanity metric. Expectancy and survival are the real ones.
FAQ
What is a good expectancy?
Any positive number means a mathematical edge. In practice, durable retail strategies often sit between +0.1R and +0.4R per trade after costs. Anything much higher is worth double-checking — it usually means costs or losing streaks have been understated.
Can I improve win rate and R:R at once?
Rarely. They tend to trade off: tighter targets raise win rate but shrink R, wider targets do the reverse. The goal is not to maximise either one but to maximise their product — expectancy — after costs.
Does this apply to options and crypto too?
Yes. Expectancy is asset-agnostic; it is pure arithmetic on your wins and losses. The instruments only change the distribution of outcomes, which is exactly why simulating matters more in fat-tailed markets.