Winning Trades ÷ Total Trades
Describes occurrence frequency only. It provides zero information about dollar or unit gain size.
Risk-Reward: Winning Often Is Not Enough, Win Size and Loss Size Must Be Weighed Together
"70% win rate" sounds impressive, and "3:1 risk-reward" sounds enticing, but reading either figure in isolation is a trap. Long-term performance requires both happening together: how frequently you win, and your average gain versus loss. Crucially, they tend to trade off against each other.
- Distinguish the distinct questions answered by win rate, risk-reward ratio, and expectancy
- Calculate per-trade mathematical expectancy using a simple formula
- Understand why an 80% win rate can lose money, while a 35% win rate can still be positive
- Recognize why win rate and payoff targets usually pull against each other
- Identify how trading costs, slippage, and sample size distort historical expectancy
Win Rate Answers "How Often"; Risk-Reward Answers "How Much on Wins vs. Losses"
Average Gain When Correct
Meaningful only when compared against average loss using an identical risk unit.
Average Loss When Wrong
Determines whether an attractive win rate will get wiped out by occasional runaway losses.
Every number in this lesson is an assumed input used to show how the arithmetic behaves. None of them is a benchmark, a target, or a figure measured from your own trading.
Expectancy = (Win Rate × Avg Win) − (Loss Rate × Avg Loss)
If 40% of trades gain +2R and 60% lose -1R, expectancy = (0.40 × 2) − (0.60 × 1) = +0.20R before costs. Here, R normalizes planned risk into a standard baseline.
3:1 Does Not Automatically Beat 1:1
Compare the two presets first: A wins 70% of the time at +1R / -1R, B wins 25% of the time at +3R / -1R. Then change the win rate and payoff assumptions yourself and watch the breakeven win rate move with them.
Expectancy matrix
3:1 does not automatically beat 1:1
Compare two presets, then change the win rate and payoff assumptions yourself. Expectancy forces probability and payoff into the same equation.
Expectancy per trade+0.4R
Breakeven win rate50%
Expectancy is positive under these assumptions.
An 80% Win Rate Can Still Bleed Capital Over Time
Suppose 80% of trades average +0.2R profit, while 20% average -1R loss: (0.80 × 0.2) − (0.20 × 1) = -0.04R. Even though 8 of 10 trades win, the 2 losses consume all prior gains and leave the total negative — before a single cent of commission.
Win Rate and Payoff Ratio Usually Pull Against Each Other
Holding the same entry rule and the same stop, pushing profit targets further out gives normal market noise more time to reach the stop first, which lowers the observed win rate. Taking profits earlier does the reverse: it raises the win rate and compresses average win size, so an ordinary stop-out costs several winners. This is a tendency inside one rule set, not a law that holds across different strategies.
Under assumed inputs, a 35% win rate with a 3R average win gives (0.35 × 3) − (0.65 × 1) = +0.40R, and a 55% win rate with a 1.2R average win gives (0.55 × 1.2) − (0.45 × 1) = +0.21R. The point is not that one archetype is better: it is that the two figures only mean something together.
| Assumed Profile | Assumed Win Rate | Assumed Avg Win / Loss | Expectancy Before Costs |
|---|---|---|---|
| Small Target, High Frequency | 75% | 0.25R win / 1.0R loss | -0.06R — (0.75 × 0.25) − (0.25 × 1); already negative before any fee |
| Mean-Reverting Range | 55% | 1.20R win / 1.0R loss | +0.21R under these inputs; costs still have to come out |
| Trend Breakout | 35% | 3.00R win / 1.0R loss | +0.40R under these inputs; the payoff size, not the win rate, carries it |
Expectancy Describes a Historical Sample, Not the Fate of Your Next Trade
Even a positive-expectancy sample can produce 5 to 10 consecutive losses in live execution. Win rates, payoffs, and transaction friction (commission, spread, slippage) also drift across market conditions, so a figure measured in one regime is not a constant.
Costs come out of every trade, not just the losers. Take the +0.21R row above and assume round-trip friction of 0.15R per trade: net expectancy falls to +0.06R, roughly a quarter of what the gross figure suggested. The same 0.15R applied to the +0.40R row leaves +0.25R. Whether a system survives its own costs depends on how large 1R is relative to the spread you actually pay.
Sample size decides how much any of this means. A 75% win rate over 8 trades is 6 wins; one different outcome moves it to 62.5%. Over 200 trades the same rate is far harder to produce by luck. Always read a win rate together with its denominator, and be explicit about which period the sample covers.
Put Your Understanding to the Test
Test your understanding of the relationship between win rate, payoff ratio, and expectancy.
Question 1 of 3
A sample shows a 50% win rate, average win of +2R, and average loss of -1R. What is the per-trade expectancy before costs?
Question 2 of 3
Why can an 80% win rate still come with negative expectancy?
Question 3 of 3
A sample shows a 35% win rate, averaging +3R on wins and -1R on losses. What can you say about it?
Stuck? Ask Mira to Break It Down
Enter your recent win rate, average gain, and average loss. Mira will calculate the expectancy those inputs imply and show the breakeven win rate once execution friction is included.
Checking sign-in status...
Put this concept into practice
Risk-Reward Ratio
Calculate a plan's risk-reward ratio, and explain why the risk-reward ratio can't be judged on its own without win rate.
Expected Value: What the Long-Run Average Trade Is Worth
Use 'expected value = win rate × average win − loss rate × average loss' to calculate the long-run average result per trade, and understand that a positive expectancy doesn't guarantee the next trade wins.