A good trading expectancy sits between 0.2 and 0.5 R: this means the trader earns on average 0.2 to 0.5 units of risk per trade over the long run. This metric, often overlooked in favor of win rate alone, is the most direct indicator of a strategy's real profitability. According to ESMA analyses covering multiple European jurisdictions, between 74 and 89% of retail CFD accounts lose money, with average losses ranging from 1,600 to 29,000 EUR per client. Most of these accounts carry strategies with negative expectancy, often without the trader realizing it. Correctly computing and interpreting your expectancy is therefore essential before committing real capital to live markets.
Trading expectancy: definition and formula
Expectancy is the quantitative foundation of any profitable trading system. It tells you, before risking a single dollar in live markets, whether a strategy is structurally profitable or not.
The exact formula
Expectancy = (Win Rate x Average Win) - (Loss Rate x Average Loss)
Expressed in R (risk multiples, assuming a fixed risk per trade):
Expectancy (R) = (Win Rate x Average R winner) - (Loss Rate x Average R loser)
Practical example with a fixed risk of $100 per trade:
- Win rate: 55%
- Average winner: $150 (1.5 R)
- Loss rate: 45%
- Average loser: $100 (1.0 R)
Expectancy = (0.55 x 1.5) - (0.45 x 1.0) = 0.825 - 0.45 = 0.375 R per trade
This means the strategy returns, on average, 0.375 times the risk per trade. Over 100 trades risking $100 each, the expected gain is $3,750.
Why express expectancy in R?
Expressing expectancy in risk multiples (R) rather than in dollars makes it independent of position size. An expectancy of 0.3 R means the same thing whether you risk $50 or $500 per trade. It is also the convention used by most prop firms to evaluate trader performance across different account sizes.
Win rate vs gain/loss ratio
Win rate and the gain/loss ratio are the two components of expectancy. They are complementary and cannot be analyzed independently.
| Win rate | R:R ratio | Expectancy | Verdict |
|---|---|---|---|
| 30% | 3:1 | +0.2 R | Profitable |
| 50% | 1:1 | 0 R | Break-even (before fees) |
| 70% | 0.5:1 | +0.2 R | Profitable |
| 80% | 0.2:1 | -0.04 R | Losing |
| 45% | 1.5:1 | +0.125 R | Marginal |
| 55% | 1.5:1 | +0.375 R | Good expectancy |
This table illustrates a common paradox: an 80% win rate strategy can carry negative expectancy if the average winner is much smaller than the average loser. Conversely, a 30% win rate strategy can be highly profitable with a strong R:R ratio.
Positive vs negative expectancy
- Positive expectancy: the strategy is theoretically profitable. It returns X R per trade on average. Capital grows over the long run, provided consistent execution.
- Zero expectancy: the strategy breaks even before fees. Broker commissions and spreads make it unprofitable in practice.
- Negative expectancy: the strategy loses money systematically. No position-sizing or money management system can save a negative-expectancy strategy indefinitely.
What is a good expectancy value?
Knowing how to calculate expectancy is only the first step. The next step is interpreting the value you obtain. Benchmarks vary by trading style.
Minimum acceptable threshold (> 0)
A positive expectancy is the necessary (but not sufficient) condition for profitability. In practice, broker fees, spreads, slippage, and execution errors all reduce your live expectancy relative to your backtest expectancy. A safety margin is therefore essential.
Practical rule: if your backtest expectancy is below 0.1 R, the strategy will likely be unprofitable live after friction costs. Target at least 0.15 R as a minimum floor.
Strong expectancy (> 0.3 R)
An expectancy above 0.3 R is considered strong by most institutional traders and prop firm evaluators. Above 0.5 R, the strategy is exceptional, and a rigorous anti-overfitting review is warranted before any live deployment.
Beware of overfitting
An expectancy above 1.0 R on a backtest is almost always a sign of overfitting. The strategy has been over-fitted to historical data and has no realistic chance of replicating that performance in live conditions. See our guide on how to backtest without overfitting to avoid this trap.
Benchmarks by trading style
Expectancy benchmarks vary significantly by trading frequency and style. The following reference values are observed in practice:
| Trading style | Typical win rate | Typical R:R | Target expectancy |
|---|---|---|---|
| Intraday scalping | 60-70% | 0.8-1.2:1 | 0.1-0.2 R |
| Day trading (SMC/ICT) | 45-55% | 1.5-2.5:1 | 0.2-0.4 R |
| Swing trading | 40-50% | 2-4:1 | 0.25-0.5 R |
| Long-term trend following | 30-40% | 4-8:1 | 0.3-0.6 R |
| Prop firm (FTMO-style) | 45-60% | 1.5-2:1 | > 0.2 R |
These are indicative ranges. An expectancy of 0.2 R is acceptable for scalping but weak for swing trading. Always interpret expectancy in the context of your specific trading style and objectives.
Calculating your expectancy on a backtest
Backtesting is the only reliable tool for computing a statistically sound expectancy. A minimum of 100 to 200 trades is required for the value to be meaningful.
Required data
To calculate expectancy on a backtest, you need:
- The complete list of all trades (winners and losers)
- The gain in dollars or R for each winning trade
- The loss in dollars or R for each losing trade
- The total number of trades
With Backtrex, this data is calculated automatically in the backtest dashboard. You get expectancy, profit factor, Sharpe ratio, and all key metrics within seconds of running a backtest. Learn how to set up your first backtest on Backtrex.
Step-by-step calculation
Count winning trades
Count losing trades
Compute rates
Apply the formula
Interpret the result
Real example: 200-trade SMC day trading backtest
Here is a sample calculation based on a 200-trade backtest on an SMC day trading strategy:
- Winning trades: 110 (win rate: 55%)
- Losing trades: 90 (loss rate: 45%)
- Average winner: 1.6 R
- Average loser: 1.0 R
Expectancy = (0.55 x 1.6) - (0.45 x 1.0) = 0.88 - 0.45 = 0.43 R
With a fixed risk of $100 per trade, this strategy generates an average of $43 per trade. Over 200 trades, the theoretical gain is $8,600. This is a strong expectancy, in the upper range for day trading strategies.
For a complete analysis, this result should be paired with profit factor, max drawdown, and Sharpe ratio. See our full guide on expectancy and profit factor metrics in backtesting.
How to improve your trading expectancy
If your backtest reveals a low or negative expectancy, three primary levers can help you improve it.
Improve your win rate
Win rate can be improved by refining entry conditions:
- Filter entries to order blocks with confluence (FVG + liquidity sweep)
- Avoid counter-trend trades against the dominant H4 or Daily trend
- Only enter during high-liquidity sessions (London and New York for Forex)
- Backtest systematically across multiple timeframes to verify the robustness of filters
Each filter should be backtested in isolation before combining them to avoid compounding overfitting effects.
Improve your R:R ratio
The risk/reward ratio can be improved by refining stops and targets:
- Use structurally-placed stops (below/above a key structure level), not arbitrary pip-based stops
- Let winning trades run with a structure-based trailing stop
- Take partial profits at key resistance/support levels
- Avoid manually closing trades before the target is reached
A higher R:R ratio reduces the required win rate: a strategy targeting 2.5:1 can be profitable with only 30% winning trades.
Cut large losses
Large losses (trades that exceed the planned stop or multiple consecutive losers) are the primary driver of expectancy destruction:
For a complete review of errors to avoid during backtesting, see our guide on common backtesting mistakes.
Limitations and pitfalls of expectancy
Expectancy is a powerful metric, but it has important limitations that every trader should understand.
Positive expectancy does not guarantee profits
A 0.3 R expectancy on 200 backtest trades does not guarantee the strategy will make money over the next six months. Variance is real: even a strong strategy can experience a run of 15 consecutive losing trades, producing a significant drawdown even with positive expectancy.
Sample size matters critically: 100 trades yield an expectancy with a wide confidence interval. 500 trades provide a significantly more reliable figure. Below 100 trades, the calculated expectancy carries no robust statistical meaning.
Psychological drawdown impact
Expectancy measures the average, not the sequence of trades. A strategy with 0.4 R expectancy may produce a run of 20 consecutive losers before recovering. Few traders sustain the psychological discipline through such a sequence. This is why expectancy must always be paired with max drawdown and maximum consecutive drawdown to assess the psychological sustainability of the strategy.
Expectancy and prop firm challenges
Prop firms such as FTMO impose drawdown rules (5% daily, 10% total) that limit available capital during losing streaks. A high-expectancy but high-drawdown strategy may fail the challenge before the expectancy has time to materialize. Optimize both expectancy and max drawdown together. See our guide on backtesting for prop firm rules for details.
Backtest expectancy vs live expectancy
Live expectancy is generally lower than backtest expectancy for several reasons:
- Slippage and spreads not perfectly modeled in the backtest
- Human execution errors (emotion, hesitation, misclicks)
- Changes in market conditions (volatility regime, correlations)
- Survivorship bias in historical data
According to ESMA analyses covering multiple European regulators, 74 to 89% of retail CFD accounts lose money, with average losses ranging from 1,600 to 29,000 EUR per client. A large share of these losses stem from deploying strategies with no prior rigorous backtesting: no expectancy calculation, no multi-year data validation.
A commonly observed ratio: live expectancy represents 60 to 80% of backtest expectancy for well-validated strategies. For unvalidated strategies, this ratio can drop below 50%. This is why rigorous multi-year backtesting on Backtrex is essential before going live.
Important Risk Warning
Conclusion: calculate, interpret, improve
Trading expectancy is the most important metric for evaluating a strategy. A value between 0.2 and 0.5 R is a solid target for most trading styles. Below 0.1 R, the strategy will likely not survive real-world friction costs.
To calculate your expectancy, run a backtest on at least 200 trades using multi-year historical data. Backtrex automatically computes expectancy, profit factor, and all key metrics. Explore our pricing plans and start validating your strategies with the rigor they deserve.
An expectancy above 0.2 R is considered acceptable. Between 0.3 and 0.5 R is good. Above 0.5 R, the strategy is excellent but warrants a rigorous anti-overfitting review. A negative expectancy, even slightly, means the strategy is structurally losing over the long term regardless of short-term results.
Expectancy = (Win Rate x Average Win) - (Loss Rate x Average Loss). Example: (55% x 1.5 R) - (45% x 1.0 R) = 0.825 - 0.45 = 0.375 R per trade. Expressing expectancy in R makes it independent of position size and comparable across strategies and account sizes.
No. Both metrics are complementary. Profit factor measures the gross ratio of total gains to total losses (ignoring win rate). Expectancy integrates both win rate and the gain/loss ratio to give the average gain per trade. Using both together provides a more complete picture of strategy quality.
Prop firms like FTMO require sufficient expectancy to reach the profit target (typically 8-10%) while respecting drawdown rules (5% daily, 10% total). In practice, an expectancy of at least 0.2 R combined with a max drawdown below 8% provides the margin of safety needed to pass the challenge reliably.
A minimum of 100 trades is needed to get an indicative expectancy figure. Between 200 and 500 trades, the value becomes statistically representative. Below 100 trades, variance is too high: you might obtain a positive expectancy purely by chance on too small a sample.
Yes. A 30% win rate strategy with a 3:1 R:R ratio has an expectancy of (0.30 x 3) - (0.70 x 1) = 0.9 - 0.7 = 0.2 R. That is a positive, acceptable expectancy. Long-term trend-following and SMC-based approaches often use this profile: fewer winning trades, but significantly larger gains relative to losses.
Live expectancy is typically 20 to 40% lower than backtest expectancy, due to slippage, real spreads, execution errors, and changing market conditions. A strategy with 0.4 R backtest expectancy may produce 0.25 to 0.32 R live. This is why targeting a sufficiently high backtest expectancy is critical to absorbing this degradation without going negative live.