A trading strategy with a 30% win rate and a 3:1 risk/reward ratio has a positive mathematical expectancy of +0.2 per trade, outperforming a strategy with a 70% win rate and a 1:3 ratio. This counterintuitive reality exposes the most common mistake among beginner traders: relying on win rate alone to evaluate their systems. According to ESMA data covering multiple European jurisdictions, between 74% and 89% of retail CFD trading accounts lose money, with average losses ranging from 1,600 to 29,000 EUR per client. Most of these strategies carry a negative expectancy, often hidden behind a win rate that looks acceptable on the surface.
Win rate: definition and common misconceptions
What win rate actually measures
Win rate measures only one thing: how often a trade closes in profit. It completely ignores the size of those profits and losses. A trader can win 9 out of 10 trades at +$50 each, then lose everything on a single -$600 trade. Their win rate is 90%, yet the strategy is a structural loser.
That gap is precisely what makes win rate misleading when used as a standalone evaluation criterion.
Why traders overestimate it
Win rate is intuitive and easy to compute. It feels like a test score: 70% sounds good, right? This analogy is dangerous in trading. Psychology also plays a major role: winning frequently provides immediate emotional satisfaction, even when individual gains are small and occasional losses are devastating.
Many trading systems sold on social media exploit this bias. They advertise win rates of 85-90% without ever mentioning the risk/reward ratio, which is often below 0.3:1. The result: each loss wipes out ten gains, and the strategy is a long-term loser.
The high win rate trap
Example: 80% win rate, losing strategy
Here is a concrete illustration. Consider four strategies evaluated over 100 trades with identical risk:
| Strategy | Win rate | Avg gain | Avg loss | Result over 100 trades |
|---|---|---|---|---|
| A | 80% | +$50 | -$300 | -$2,000 (loser) |
| B | 40% | +$300 | -$100 | +$6,000 (winner) |
| C | 30% | +$400 | -$100 | +$5,000 (winner) |
| D | 55% | +$150 | -$100 | +$3,750 (winner) |
Strategy A shows an 80% win rate (4 trades out of 5 profitable). Yet over 100 trades: (80 x 50) - (20 x 300) = 4,000 - 6,000 = -$2,000. Strategy B, with only 40% win rate, generates: (40 x 300) - (60 x 100) = 12,000 - 6,000 = +$6,000.
This illustrates the core paradox: a high win rate does not guarantee profitability. Only the combination of win rate and position sizing (gain-to-loss ratio) determines the actual outcome.
The role of the average gain-to-loss ratio
The risk/reward ratio is the missing piece that win rate ignores. It expresses how much a winning trade returns on average compared to what a losing trade costs.
A ratio of 2:1 means gains are twice as large as losses. With that ratio, a win rate of just 40% is enough to be profitable: (0.4 x 2) - (0.6 x 1) = 0.8 - 0.6 = +0.2 per trade.
Breakeven win rate by risk/reward ratio
To break even with a 3:1 ratio, a win rate of just 26% is sufficient. With a 1:1 ratio, you need above 50%. With a 0.5:1 ratio, you need above 67%. The exact formula is: Minimum win rate = 1 / (1 + R:R ratio).
The metrics that actually matter
Mathematical expectancy
Mathematical expectancy is the metric that combines win rate and risk/reward ratio into a single indicator. It represents the average expected gain per trade, expressed in multiples of the risk taken (R).
Formula: Expectancy = (Win Rate x Avg gain) - (Loss Rate x Avg loss)
Positive expectancy means the strategy is structurally profitable over the long run. Zero or negative expectancy leads to losses regardless of position sizing or trader psychology.
Generally accepted benchmarks among professional traders:
- Expectancy above 0.3 R per trade: excellent strategy
- Expectancy between 0.1 and 0.3 R: good strategy
- Expectancy between 0 and 0.1 R: marginal (trading costs can push it negative)
- Expectancy below 0: reject the strategy without exception
For a deeper dive, see our complete guide on trading expectancy and our article on expectancy and profit factor in backtesting.
Profit factor
Profit factor is the ratio of total gains to total losses. Formula: Profit Factor = Sum of gains / Sum of losses.
A profit factor of 1.0 means the strategy breaks even. A profit factor of 1.5 means that for every dollar lost, the strategy generates $1.50. This is the minimum recommended threshold before going live.
Reference thresholds:
- Profit factor above 2: robust strategy
- Profit factor between 1.5 and 2: acceptable strategy
- Profit factor between 1 and 1.5: fragile strategy, likely to degrade out-of-sample
- Profit factor below 1: losing strategy
The advantage of profit factor over win rate: it accounts for every dollar gained and lost, not just the count of trades.
Sharpe ratio and Sortino ratio
The Sharpe ratio measures risk-adjusted return. It expresses how many units of return are obtained per unit of risk taken. Formula: Sharpe = (Average return - Risk-free rate) / Standard deviation of returns.
Academic research on the Sharpe ratio shows that Berkshire Hathaway achieved a Sharpe ratio of 0.79 over the period 1976-2017, while the US stock market reached 0.49 over the same period. In active trading, a Sharpe ratio above 1 is generally considered acceptable, and above 2 is excellent.
The Sortino ratio improves on the Sharpe by penalizing only downside volatility (losses), rather than all volatility. It is particularly suited to asymmetric strategies where gains can occasionally be outsized.
For a detailed breakdown, see our guide on Sharpe, Sortino, and Calmar ratios in backtesting.
How to backtest using the right metrics
Essential metrics dashboard
A complete backtest report should include the following metrics to allow an honest strategy evaluation:
| Metric | What it measures | Acceptable threshold |
|---|---|---|
| Win rate | Frequency of winning trades | Depends on R:R (useless alone) |
| Avg risk/reward ratio | Avg gain / avg loss ratio | Above 1:1 minimum |
| Expectancy (R) | Average gain per trade in risk multiples | Above 0.1 R |
| Profit factor | Total gains / total losses ratio | Above 1.5 |
| Sharpe ratio | Risk-adjusted return | Above 1.0 |
| Max drawdown | Worst consecutive loss from a peak | Below 20-25% of capital |
| Trade count | Statistical volume of the sample | Above 200 trades |
Trade count is often overlooked: a backtest on 20 trades can display any statistic without being statistically meaningful. Recommended minimum: 200 trades over the tested period.
To understand how to read a full backtest report, see our guide on reading backtesting results and metrics.
Backtesting with Backtrex
Most backtesting platforms display win rate first, sometimes as the only metric. This display bias directly contributes to poor strategy evaluation by inexperienced traders.
Backtrex takes a different approach: every backtest report automatically shows 12 performance metrics, including expectancy, profit factor, Sharpe ratio, and Sortino ratio. Win rate is present but contextualized alongside the other indicators to prevent misinterpretation.
The no-code platform lets you test any strategy on historical data in under 30 seconds, without writing a single line of code. Explore Backtrex's backtesting features or review our pricing plans.
Avoid confirmation bias in backtesting
Before running a backtest, define your acceptance criteria upfront: for example, "expectancy above 0.15 R AND profit factor above 1.5 AND drawdown below 20%." This prevents adjusting criteria after the fact to make results look better than they are, a common bias in backtesting known as data snooping.
To go deeper on the relationship between risk/reward and expectancy, see our article on risk/reward ratio and backtesting optimization.
Important Risk Warning
Conclusion
Win rate is an appealing metric for its simplicity, but dangerously incomplete when analyzed in isolation. A profitable strategy is evaluated across all its indicators: expectancy, profit factor, Sharpe ratio, and maximum drawdown. Only a complete backtesting analysis can objectively determine whether a strategy deserves live capital.
Yes, absolutely. If the average gain-to-loss ratio exceeds 1.5:1, a 40% win rate generates a positive expectancy of +0.2 R per trade. Over 200 trades with $100 risk per trade, that represents a theoretical gain of $4,000. Many professional traders and algorithmic systems operate with win rates below 50% by maintaining high risk/reward ratios.
Win rate measures the percentage of winning trades, ignoring their size. Profit factor is the ratio of total gains to total losses: it accounts for both the frequency and the magnitude of wins and losses. A profit factor of 1.5 means the strategy generates $1.50 for every dollar lost, regardless of win rate.
There is no universal minimum win rate: it all depends on the risk/reward ratio. With a 3:1 ratio, a 26% win rate is enough to break even. With a 1:1 ratio, you need above 50% before costs. With a 0.5:1 ratio, you need above 67%. The exact formula is: Minimum win rate = 1 / (1 + R:R ratio).
Win rate ignores the size of gains and losses. A strategy can win 9 out of 10 trades, but if the single loss is 20 times larger than each gain, the strategy loses money over time. Complete strategy evaluation requires at minimum: win rate combined with risk/reward ratio, or directly the expectancy and profit factor from a backtest.
Yes, the Sharpe ratio measures risk-adjusted return. A ratio above 1 indicates the strategy generates more return than volatility, which is a good signal. However, it has important limitations: it assumes normally distributed returns, which is not always the case in active trading. The Sortino ratio, which penalizes only downside volatility, is often preferred for active strategies.
The formula is: Expectancy = (Win Rate x Avg gain) - (Loss Rate x Avg loss). Example: 45% win rate, average gain $200, average loss $100. Expectancy = (0.45 x 200) - (0.55 x 100) = 90 - 55 = $35 per trade. Expressed in R: Expectancy (R) = (0.45 x 2) - (0.55 x 1) = 0.9 - 0.55 = +0.35 R per trade.