Risk-Reward, Win Rate and Expectancy

"Tags: risk reward, win rate, expectancy, R multiples Prerequisites: Position Sizing; Risk Units A strategy is not defined by how often it wins or by one attractive reward ratio. It is defined by the relationship between all wins, all losses and their frequency. Questions this article answers What does risk-reward ratio actually measure? Why is a high reward ratio not automatically better? How do win rate, average win and average loss interact? What is expectancy? Why must realised results replace imagined targets? 1. Planned risk-reward Risk-reward compares the planned potential gain with the planned loss. If the planned risk is one unit and the potential gain is two units, the planned ratio is 2:1. The ratio is a scenario, not a guarantee that price will reach the reward level. 2. Realised risk-reward Realised risk-reward uses the actual gain or loss after the trade closes. A trade planned for +3R may exit at +0.8R, -1R or +5R. Strategy evaluation must use realised results rather than attractive pre-trade projections. 3. Win rate Win rate is the proportion of closed trades that produce a positive result. A high win rate can coexist with poor profitability if losses are much larger than wins. A low win rate can be profitable if winners are sufficiently large and losses remain controlled. 4. Average win and average loss Average win is the mean positive outcome across winning trades. Average loss is the mean negative outcome across losing trades. Together with win rate, these values describe the basic payoff structure. 5. Expectancy concept Expectancy is the average amount a strategy can be expected to gain or lose per trade over a large sample, based on its historical distribution. Conceptually, it combines the probability and size of wins with the probability and size of losses. Positive expectancy does not mean every trade, week or month will be profitable. 6. Illustrative expectancy examples 7. Why high reward targets can reduce win rate A distant target offers a larger theoretical reward but is reached less often. A closer target is reached more frequently but produces smaller wins. The strategy must balance payoff size and probability through evidence, not preference. 8. Why tight stops distort reward ratios A very tight stop can create a large-looking reward ratio on paper. If the stop is inside normal volatility, the strategy may suffer frequent losses before the expected move develops. Logical invalidation comes before ratio optimisation. 9. Open-ended winners Trend-following methods may not use a fixed target. They can allow winners to continue while trailing risk. The realised reward distribution is then evaluated through trade records rather than a predeclared target. 10. Partial profits and expectancy Partial exits can improve psychological comfort and reduce exposure. They also reduce the size of large winners if too much is sold early. Their effect should be evaluated across the full strategy distribution. 11. Breakeven exits A breakeven exit avoids a loss but is not free of consequence. Moving stops too quickly can convert future winners into zero outcomes. The frequency of breakeven trades and their opportunity cost should be recorded. 12. Trading costs Brokerage, taxes, exchange fees, spread and slippage reduce realised expectancy. High-frequency or low-reward strategies are more sensitive to costs. Gross profitability should not be confused with net profitability. 13. Sample size A small sample can create misleading win rates and averages. A few large winners or losses can dominate early results. Expectancy should be reviewed over a meaningful number of comparable trades and across different market regimes. 14. Distribution matters Two strategies can have the same average expectancy but very different experiences. One may produce steady small gains; another may rely on rare large winners and long losing streaks. The trader must understand not only the average but also the variability. 15. Maximum adverse excursion Maximum adverse excursion records how far a trade moved against the position before closing. It can help evaluate whether stops are inside normal fluctuation or unnecessarily wide. The metric should be analysed across many trades, not used to justify widening one current stop. 16. Maximum favourable excursion Maximum favourable excursion records the best unrealised progress during a trade. It can reveal whether exits regularly surrender large gains or whether targets are unrealistic. It supports exit research but does not predict the next trade. 17. Expectancy and market regime A breakout strategy may perform differently in a trending market and a choppy market. Overall expectancy can hide regime-specific weakness. Results should be segmented by setup type, market condition and execution quality. 18. Expectancy worksheet 19. Common beginner mistakes Judging a strategy by win rate alone Payoff size can reverse the conclusion. Using imagined targets as realised reward Evaluation must use actual exits. Forcing a 3:1 ratio on every chart Market structure may not support the target. Ignoring costs and slippage Net expectancy can be much lower. Changing the method after ten trades The sample may be too small. Ignoring regime differences A strategy can be positive overall but fail in certain conditions. 20. DStreet principle A strategy is not good because it wins often or promises large targets. It is good only if the complete, realised distribution remains positive and survivable. 21. Beginner checklist Planned reward is not realised reward. Win rate must be combined with average win and average loss. Tight stops can create misleading reward ratios. Costs reduce expectancy. Large samples are more reliable than short streaks. Distribution and drawdown matter alongside the average. Results should be segmented by market regime and setup. 22. Quick knowledge check Question: Can a 70% win-rate strategy lose money? Answer: Yes, if losses are much larger than wins. Question: Can a 40% win-rate strategy be profitable? Answer: Yes, if average wins sufficiently exceed losses. Question: What is expectancy? Answer: The average outcome per trade across a large sample. Question: Why is a planned 3R target insufficient? Answer: The market may not reach it and actual exits differ. Question: Why does sample size matter? Answer: Small samples can be dominated by randomness."