What is Payoff Ratio?
Reviewed by Karthikeyan Anbarasan (Founder, TradeLyser) · Updated 19 Oct 2026 · Editorial policy
How the payoff ratio is calculated
- Take the closed trades for one setup over a period, net of charges.
- Split them into wins and losses. Decide once how to treat scratch trades near zero, and keep to it.
- Average the wins: total won ÷ number of wins.
- Average the losses as a positive number: total lost ÷ number of losses.
- Divide the average win by the average loss.
Worked example: 50 Bank Nifty option trades
Illustrative, after charges. 22 wins totalled ₹66,000 and 28 losses totalled ₹42,000. Expectancy is 0.44 × ₹3,000 − 0.56 × ₹1,500 = ₹1,320 − ₹840 = ₹480 per trade, or ₹24,000 over 50 trades, which matches ₹66,000 − ₹42,000. The breakeven win rate at a payoff of 2.0 is 33.3%, so the setup has 10.7 points of room. Profit factor is ₹66,000 ÷ ₹42,000 = 1.57, which is the payoff ratio × wins ÷ losses: 2.0 × 22 ÷ 28. Now suppose the same trader starts booking profits early and the average win drops to ₹2,000 while the win rate stays at 44%. The payoff ratio falls to 1.33, the breakeven win rate rises to 42.9%, and expectancy shrinks to 0.44 × ₹2,000 − 0.56 × ₹1,500 = ₹40 a trade, almost nothing once a few extra charges are counted.
Payoff ratio and win rate together
Payoff ratio
Risk-reward ratio
Profit factor
Using it, and where it misleads
The payoff ratio is most useful as a trend. Work it out per setup each month and put it next to the win rate. If the payoff ratio falls while the win rate rises, exits have probably become quicker; if the payoff ratio rises while the win rate falls, stops may have become tighter or targets further away. Neither is wrong in itself. What matters is whether expectancy, which combines the two, moved up or down.
- Compare planned risk-reward with realised payoff each month. If you plan 1:2 and realise 1.2, exits are cutting winners short or stops are slipping.
- Cutting winners early raises win rate and lowers payoff; widening stops does the same to losses. Watch both numbers move together.
- One outsized win can lift the payoff ratio for months. Recompute without the largest trade.
- Keep it per setup. A scalping setup and a swing setup mixed together give a payoff that describes neither.
- Below about 30 trades per setup, the figure moves a lot with each new trade. Quote it with the trade count, for example "payoff 2.0 over 50 trades".
- Use net figures. Charges make every win a little smaller and every loss a little larger, so a gross payoff ratio always looks better than the real one.
Reference guide
| Context | Value | Reading |
|---|---|---|
| Payoff ratio read together with win rate | Complete picture | A win rate above 1 ÷ (1 + payoff) means positive expectancy before charges |
| High payoff driven by one very large win | Fragile | Remove the largest trade and recompute; check the median win as well |
| Payoff below 1 and nobody checks the win rate | Risky | It needs a win rate above 50%, and a few large losses can erase months of small wins |
Questions
What is a good payoff ratio?
It depends on your win rate. Any payoff works if the win rate is above 1 ÷ (1 + payoff): a payoff of 2.0 needs more than 33.3%, a payoff of 1.0 needs more than 50%.
How do you calculate the payoff ratio?
Divide the average winning trade by the average losing trade, both in rupees or both in R, using the same set of closed trades. ₹3,000 ÷ ₹1,500 = 2.0.
Is the payoff ratio the same as the risk-reward ratio?
No. Risk-reward is planned for one trade before entry; the payoff ratio is what your wins and losses actually averaged over many trades. The gap between the two shows how your exits and stops behave.
What is the difference between payoff ratio and profit factor?
The payoff ratio compares average sizes and ignores how often you win. Profit factor compares totals, so it includes win rate: profit factor = payoff ratio × number of wins ÷ number of losses.
Can a strategy with a payoff ratio below 1 be profitable?
Yes, if its win rate is high enough. A payoff of 0.8 breaks even at about 55.6% wins, so it needs a win rate above that after charges.