Profit factor is the gross profit of all winning trades divided by the gross loss of all losing trades, taken as a positive number, over the same set of trades. A value above 1 means the trades won more in total than they lost; a value below 1 means they lost money overall.
How it works
The formula is simple: profit factor = gross profit ÷ gross loss. A set of trades that won $15,000 and lost $10,000 has a profit factor of 1.5: each dollar lost was matched by $1.50 won. Two equivalent forms show what drives it:
- From win rate and payoff. With a win rate W and a payoff ratio P (average win ÷ average loss), profit factor = (W × P) ÷ (1 − W).
- From net profit. Net profit = gross loss × (profit factor − 1). At a profit factor of 1.25, the net profit equals a quarter of everything lost along the way.
Profit factor is closely tied to trading expectancy: expectancy is positive exactly when the profit factor is above 1. The two answer different questions, though. Expectancy says how much a trade earns on average, in dollars or R; profit factor is a ratio without units, so it compares systems of any size but says nothing about how much each trade earns, how often the system trades or how deep its drawdowns run.
How to read it
Profit factor before costs for a given win rate and payoff ratio (rounded to two decimals):
| Win rate | Payoff 1 | Payoff 1.5 | Payoff 2 | Payoff 3 |
|---|---|---|---|---|
| 30% | 0.43 | 0.64 | 0.86 | 1.29 |
| 40% | 0.67 | 1.00 | 1.33 | 2.00 |
| 50% | 1.00 | 1.50 | 2.00 | 3.00 |
| 60% | 1.50 | 2.25 | 3.00 | 4.50 |
The same profit factor can come from very different systems: a trend method that wins 40% of the time with a payoff of 2 and a mean-reversion method that wins 57% of the time with a payoff of 1 both sit near 1.33. What counts as a good value is a matter of rough convention rather than a fixed rule. Commonly cited ranges for a tested rule set with many trades, after costs:
- Below 1: the trades lost money.
- 1 to about 1.3: a thin margin that commissions, slippage or a change in conditions can erase.
- About 1.5 to 2: often treated as a workable margin.
- Well above 3: worth checking for a small sample, one outsized winner or a rule fitted too closely to past data.
Worked example (illustrative numbers)
A journal holds 40 closed trades of one setup: 16 winners averaging $600 and 24 losers averaging $250.
| Check | Calculation | Profit factor |
|---|---|---|
| As recorded | $9,600 ÷ $6,000 | 1.60 |
| After costs of $10 per round trip | (16 × $590) ÷ (24 × $260) = $9,440 ÷ $6,240 | about 1.51 |
| Largest winner ($2,400) removed | $7,200 ÷ $6,000 | 1.20 |
| Square-root adjustment | (16 − 4) × $600 ÷ ((24 + 4.9) × $250) | about 1.00 |
The last row is a conservative check that some backtesting platforms report as an adjusted profit factor: it removes the square root of the number of winners from the winning trades and adds the square root of the number of losers to the losing ones, a rough allowance for luck in a small sample. With 40 trades, a recorded 1.60 shrinks to break-even. The trades may still reflect a real edge, but this sample cannot show it yet.
Limitations and common mistakes
- Small samples. A handful of trades can produce almost any value. Judge profit factor over many trades, and per setup rather than mixed.
- Outliers. One very large winner can carry the whole figure; recalculate without it.
- Leaving out costs or the trades that "did not count", which overstates the result.
- Open losers. Profit factor uses closed trades only, so a losing position held open instead of stopped out flatters it.
- Ignoring the path. Two systems with the same profit factor can have very different losing streaks and maximum drawdowns.
- Optimizing for it. Picking the parameters with the highest profit factor in a backtest is a direct route to overfitting.
- No losing trades. With zero gross loss the ratio is undefined, not infinite proof of an edge.
In Clear Tape
The free trade journal shows profit factor (gross profit ÷ gross loss) next to win rate, expectancy in $ and R and maximum drawdown, calculated from your own closed trades; until a losing trade is recorded it says so instead of showing a number. The free expectancy calculator returns the profit factor implied by a win rate, an average win and an average loss. Backtests on the Systems page report win rate, profit factor and expectancy after costs on past prices; they are available to Tools and Pro members. More about the journal is on the trade journal page.
For educational purposes only. Not investment advice.