Win rate is not the score
2026-10-11 · 3 min read
The first metric most people examine is win rate: how often a strategy is correct. It is also one of the least informative metrics on its own. A method that loses more often than it wins can still be profitable, and one that wins most of the time can still lose money. The outcome depends on the relative size of average wins and losses.
A simple example
Consider units of risk. The safety exit defines the loss on a single trade; that amount is 1R, or one unit of risk. A win is measured in the same units: did the trade earn 2R or 0.5R before it closed? Now compare two strategies:
| Strategy | Win rate | Average win | Average loss | Per trade |
|---|---|---|---|---|
| A | 40% | +2.0R | −1.0R | +0.20R |
| B | 70% | +0.5R | −1.0R | +0.05R |
Strategy B is correct far more often, yet Strategy A earns four times as much per trade. The calculation is the sum of wins and losses: 0.40 × 2.0 − 0.60 × 1.0 = +0.20R. That total, the average return per trade, is called expectancy, and it is what compounds over time.
The break-even win rate
The required win rate depends on the size of average wins relative to average losses. That ratio is the payoff (average win ? average loss). The win rate required to break even is 1 ÷ (1 + payoff):
| Payoff (win ÷ loss) | Win rate needed to break even |
|---|---|
| 3:1 | 25% |
| 2:1 | 33% |
| 1:1 | 50% |
| 0.5:1 | 67% |
A strategy that lets winners run can be profitable even when it is wrong most of the time, while a strategy that takes small profits must be right almost always. Neither is inherently better; they represent different assumptions about how prices move.
The trade-off between win rate and payoff
Widening the profit target to raise the payoff is not a free adjustment. A more distant target is reached less often, so the win rate falls. Conversely, taking profits sooner raises the win rate but reduces the average win. Moving the target changes both metrics; what matters is their product, not either metric alone.
How this appears in the record
Because win rate alone can mislead, the published record pairs it with payoff and displays the break-even win rate next to the actual rate. If the actual win rate is above break-even, the method is profitable before costs; if it is below, the average win is too small to offset the losses. It is a more accurate measure than the headline percentage.
Important caveats
- Expectancy is meaningful only over many trades. Over a small sample, luck dominates; a high win rate over ten trades conveys little.
- Costs reduce returns. Fees and slippage reduce every win and increase every loss, so a thin edge can disappear before it is realized.
- Positive expectancy still allows losing streaks. Being correct on average is not the same as being correct today.
This is research and education, not investment advice. Trading cryptocurrency involves substantial risk of loss.
See the win rate and the payoff together
The public record is free and read-only: daily close results, drawdowns, and methodology, with no sign-in required.