Win Rate
Win rate is the number of profitable trades divided by the total number of trades, expressed as a percentage. A strategy with 60 winners out of 100 trades has a 60% win rate. The number is intuitive but never sufficient on its own — a high win rate with a poor payoff ratio is a losing strategy.
The win-rate trap
A 90% win rate sounds impressive. But if your average win is $10 and your average loss is $200, you have a losing system: 0.9 × $10 = $9 per winning trade against 0.1 × $200 = $20 per losing trade — net expectancy −$11. Strategies marketed on win rate alone are almost always martingale-style averaging-down systems that look perfect until the one bad trade wipes the account.
The fix: always read win rate together with the payoff ratio (average win / average loss).
Breakeven win rate by payoff ratio
The mathematical condition for profitability is:
win_rate × avg_win > (1 − win_rate) × avg_loss
Solving for the breakeven win rate at different payoff ratios:
| Payoff (avg win ÷ avg loss) | Breakeven win rate |
|---|---|
| 0.5 (wins half size of losses) | 67% |
| 1.0 (wins same as losses) | 50% |
| 1.5 | 40% |
| 2.0 | 33% |
| 3.0 | 25% |
| 5.0 | 17% |
Above the breakeven row your strategy makes money; below it loses. A strategy with 30% win rate at 3:1 payoff is more profitable than 65% win rate at 0.5 payoff.
Win rate by strategy class
Different strategy classes naturally cluster at different win-rate ranges. Knowing the typical band for your style stops you from over-engineering toward an unrealistic number.
- Trend following — 30-45% win rate, 2.5-4x payoff. Most trades stop out small; the few that catch big moves carry the year.
- Breakout — 35-45% win rate, 2-3x payoff. Similar profile, slightly higher win rate because trend confirmation filters reduce false starts.
- Mean reversion — 55-70% win rate, 0.7-1.2x payoff. Many small wins as price snaps back to the mean, occasional large losses when it doesn't.
- Scalping / range trading — 60-75% win rate, 0.5-0.9x payoff. Profit factor depends on execution costs.
- Options selling (premium-collection) — 80-90% win rate, 0.1-0.3x payoff. Looks beautiful in equity curve but one tail event eats years of premium.
If your trend-following backtest shows a 70% win rate, double-check for look-ahead bias. It almost certainly has one.
Why traders chase win rate (and shouldn't)
A high win rate feels good. Every trade that closes green reinforces the strategy. Long strings of wins create false confidence; long strings of losses (normal in a 35% win-rate trend system) create false despair. Most retail traders abandon perfectly good trend strategies during a routine losing streak — not because the math broke, but because their psychology did.
Two ways to inoculate against this:
- Plan for the maximum streak. A 35% win-rate strategy will see 10 consecutive losses about once every 240 trades. Knowing this in advance makes it survivable.
- Track expectancy, not streaks. Expectancy = (win_rate × avg_win) − ((1 − win_rate) × avg_loss). Positive expectancy + adequate sample size = profitable. Streaks are noise.
Improving win rate without breaking the strategy
Adding filters that remove low-probability setups is the cleanest way. The trick: every filter you add removes losing trades *and* winning trades. You want filters that remove proportionally more losers than winners.
- Trend regime filter (e.g. only trade longs above the 200 EMA) — removes counter-trend setups, typically lifts win rate by 5-10 percentage points.
- Volatility filter (only trade when ATR is rising) — improves win rate on breakouts.
- Time-of-day filter — many forex pairs mean-revert in Asian hours and trend in NY hours. Restricting to one regime cleans up the win-rate distribution.
Avoid adding filters that just memorise past unfavourable conditions ("don't trade Tuesdays in March"). They lift in-sample win rate and fail in walk-forward.
Win rate and position sizing
Win rate determines the maximum bet size you can survive. The Kelly criterion formalises this:
Kelly fraction = win_rate − ((1 − win_rate) / payoff)
For a 50% win-rate strategy with 1.5x payoff: 0.5 − (0.5 / 1.5) = 0.17. Full Kelly says risk 17% of equity per trade — way too high in practice. Most professionals use fractional Kelly (1/4 to 1/2 of the formula's output), which is why "risk 1-2% per trade" is the retail rule of thumb.
A low win rate (30-40%) tolerates much smaller per-trade risk than a high win rate at the same payoff, simply because long losing streaks are more likely.
Win rate in PineForge
Every PineForge backtest reports win rate alongside profit factor, payoff ratio, and trade count. Use the trio together — never one number in isolation. See our risk management guide for how win rate flows into position sizing.
Frequently Asked Questions
What is a good win rate for a trading strategy?+
There is no single number — it depends on payoff ratio. A 30% win rate with 3:1 payoff outperforms a 65% win rate with 0.5 payoff. The right question is whether your win rate is above the breakeven for your payoff ratio: above 33% with 2:1 payoff, above 50% with 1:1 payoff, above 67% with 1:2 payoff.
Why do trend-following strategies have low win rates?+
Trend strategies stop out quickly on every false start, then ride the rare real trends to outsized winners. The structure is many small losses and a few big wins by design. A trend-following strategy with a 70% win rate is almost always overfit or has look-ahead bias — the math of trend distributions makes that win rate impossible without changing the strategy class.
Can I be profitable with a 30% win rate?+
Yes — most professional trend-followers and CTAs run at 30-45% win rates. The key is the payoff ratio: average win must be at least 2.5-3x average loss. That requires letting winners run with trailing stops while cutting losers fast — psychologically hard but mathematically straightforward. The 30% win rate is the trend-follower's edge, not their handicap.
How does win rate affect position sizing?+
Lower win rates require smaller position sizes because long losing streaks become more likely. The Kelly criterion captures this: optimal bet fraction = win_rate − ((1 − win_rate) / payoff). Most retail traders should use fractional Kelly (1/4 to 1/2 of the formula) and cap risk at 1-2% per trade — the rule of thumb that keeps you alive through statistically inevitable losing streaks.
How many trades do I need for a win rate to be statistically meaningful?+
At least 100 trades for a rough estimate, 300+ for confidence. With only 30 trades a win rate of 60% has roughly ±18 percentage points of uncertainty — meaning the true win rate could be anywhere from 42% to 78%. Always check trade count alongside the headline number.
Is a high win rate a sign of overfitting?+
Often, yes. If a backtest shows 80%+ win rate on a strategy class that normally runs at 40-50%, the most likely explanation is overfitting or look-ahead bias. Run walk-forward analysis on the strategy. If the out-of-sample win rate drops by more than 15 percentage points, the in-sample number was a fantasy.
Related Terms
Profit Factor
Profit factor is gross winning trades divided by gross losing trades — the single quickest test of whether a strategy has any real edge at all.
Sharpe Ratio
The Sharpe ratio measures excess return per unit of volatility — the most-cited risk-adjusted performance metric in finance.
Drawdown
Drawdown is the percentage decline from a peak in your account equity to the next trough — the most important number in risk management.
Stop Reading. Start Trading.
PineForge backtests every concept on this site against real market data.
Try It Free