Lesson 35 min read
Risk-to-reward and expectancy: why win rate is not enough
How a trader can be wrong more often than right and still come out ahead, and how to check the numbers yourself.
Measuring in R
Call the amount you risk on a trade 1R. If you risk $50, then 1R is $50. A trade that makes $100 made 2R. A trade that reaches its stop lost 1R. Thinking in R lets you compare trades of any size.
Risk-to-reward
Risk-to-reward compares what you risk with what you aim to make. Risking 1R to aim for 2R is written 1:2.
Why win rate alone tells you little
Take two traders over ten trades each. Trader A wins 4 of the 10. Wins make 2R and losses cost 1R.
4 × 2R − 6 × 1R = +2R
Trader B wins 7 of the 10. Wins make 0.3R and losses cost 1R.
7 × 0.3R − 3 × 1R = −0.9R
Trader A is right less than half the time and ends ahead. Trader B is right most of the time and ends behind.
Expectancy
Expectancy is the average result per trade over many trades:
Expectancy = (win rate × average win) − (loss rate × average loss)
For Trader A:
0.4 × 2R − 0.6 × 1R = +0.2R
A positive number means the method made money over that sample. A negative number means it lost.
The win rate you need
Each risk-to-reward has a break-even win rate, before costs:
- 1:1 needs 50%.
- 1:2 needs about 33%.
- 1:3 needs 25%.
What the numbers leave out
- Costs such as spreads and commissions come out of every trade and lower expectancy.
- A few trades prove nothing. You need a large sample before the average means much.
- Past results do not tell you what the next trades will do.
Key takeaways
- Judge a method by its expectancy, not by its win rate.
- A high win rate with small wins and large losses can still lose money.
- Count the costs, and do not trust a small sample.