Win rate, RR and profit factor: how to calculate them
Win rate is the first number everyone calculates and the least useful of the three. Here is how they relate, and which combination actually tells you whether a system earns.
Short version. Win rate without RR means nothing. A 30% win rate at an average RR of 3 is profitable trading; a 70% win rate at RR 0.4 is losing trading. The two numbers that matter are profit factor (all profit ÷ all loss) and expectancy (the average result of one trade).
Win rate
Win rate is the share of winners among all closed trades. 40 winners out of 100 trades is 40%.
The trouble is that win rate knows nothing about size. Forty trades at +0.3% and sixty at −1% give a 40% win rate and a total of −48% in risk terms. The same 40%, with winners at +3% and losers at −1%, gives +60%. Identical number, opposite outcome.
The one useful role of win rate is as an internal check: if it dropped while your entry rules didn't change, then either the market changed or you did.
RR
RR is how many times the profit on a trade exceeds the risk. Risk 1%, take 2.5%, and RR is 2.5.
Measure RR from the stop and the target rather than from the outcome: it describes the entry, not your luck. And measure what happened, not what was planned — "planned 1:4, took 1:1.2" is RR 1.2, whatever was drawn on the chart.
Win rate and RR are linked by a simple equation. To be profitable you need:
win rate > 1 / (1 + RR)
| Average RR | Minimum win rate |
|---|---|
| 1.0 | 50% |
| 1.5 | 40% |
| 2.0 | 33% |
| 3.0 | 25% |
| 4.0 | 20% |
That is the breakeven threshold before commissions and spread. The real one sits slightly higher.
Profit factor
Profit factor is all profit divided by all loss. Winners brought 18%, losers took 9% — profit factor 2.0.
It is more honest than win rate because it accounts for size, and more honest than average RR because it accounts for frequency. Rough markers: below 1 the system loses money; 1.0–1.3 is around breakeven and any run of stops knocks it over; 1.5 and up is workable; above 3 on a small sample almost always means one or two trades made the result.
That last point is the main trap. Profit factor over 15 trades means nothing — a single large trade multiplies it. Look at it from 50–100 trades, and it is worth checking what happens if you remove your best trade. If profit factor falls below 1 without it, there is no edge yet, only luck.
Expectancy
Expectancy answers the question directly: how much one of your trades earns on average.
expectancy = win rate × average win − loss rate × average loss
Example: win rate 45%, average win 1.9R, average loss 1R. Expectancy = 0.45 × 1.9 − 0.55 × 1 = +0.31R per trade. At 60 trades a month and 1% risk that is roughly +18% a month — over a long run, not every month.
What makes expectancy practical is that it translates into a plan: want more, and you either trade more often, at higher RR, or with fewer junk entries. The third is usually cheaper than the first two.
What to calculate first
- Profit factor across the whole journal — does the trading work at all.
- Profit factor per setup — what exactly works. There is almost always one setup dragging the result down that you keep taking out of habit.
- Expectancy — what that is worth in percent per trade.
- Drawdown — whether you can take that trading psychologically, and whether it fits your prop firm's limits.
Win rate isn't on the list: it's already inside the other three.
Where this gets calculated for you
StatsAI computes all four — plus breakdowns by setup, instrument and session — as you log trades. In a spreadsheet or in Notion the same thing needs formulas, and drawdown along the equity curve can't properly be done there at all: it needs a running total in trade order.
updated October 27, 2026