ARMFProfit factor
The profit factor is what was earned on winning trades divided by what was lost on losing ones over a period. One number that shows how many times more the system earns than it loses. Convenient for comparison but easily misleading on a short sample and with uneven trades.
What profit factor is in trading: the formula and the meaning
profit factor = earned on winning trades ÷ lost on losing ones
A value of 1.0 means break-even: what was earned equals what was lost. 1.5 means a dollar and a half of profit for every dollar of losses. Below 1.0 the system loses.
| Profit factor | How to read it | Caution |
|---|---|---|
| Below 1.0 | The system lost over the period | Check whether the costs are the reason |
| 1.0–1.2 | Barely positive | Sensitive to a widening spread and to slippage |
| 1.3–1.8 | The working range | Normal for a system with real costs |
| 1.8–2.5 | A high result | Check the number of trades and the uniformity of the period |
| Above 2.5 | Suspicious on a small sample | Often the consequence of one or two large trades |
The upper rows call not for admiration but for a check. A profit factor of 3.0 over thirty trades almost always means one lucky trade made the whole statistic: remove it and the value drops below 1.5.
The link with the win rate and the ratio
The profit factor is not an independent metric: it is derived from the win rate and the average risk-to-reward ratio.
profit factor = (win rate × average R) ÷ (1 − win rate)
This is a useful check on the record-keeping. Calculate the profit factor directly from the sums and through the win rate formula: a discrepancy means the averages are distorted by outliers — for example by one trade closed far beyond the target.
| Win rate | Average R | Profit factor | Expectancy, R |
|---|---|---|---|
| 35 % | 2.0 | 1.08 | +0.05 |
| 40 % | 2.0 | 1.33 | +0.20 |
| 45 % | 2.0 | 1.64 | +0.35 |
| 50 % | 2.0 | 2.00 | +0.50 |
| 45 % | 1.5 | 1.23 | +0.13 |
| 45 % | 3.0 | 2.45 | +0.80 |
Three situations where the profit factor deceives
Checking a forex system's profit factor for robustness
One number rarely lies by itself — what lies is the sample it was calculated on. Four checks take ten minutes and filter out most false conclusions.
Recalculate the profit factor without it. A fall below 1.2 means the result was made by one hit rather than by a system.
check 1Calculate separately for the first and the second half. A twofold discrepancy is a sign that two different market regimes have been averaged.
check 2Often the overall figure rests on one pair while the rest are in the red. That is a reason to narrow the list rather than to change the rules.
check 3If the statistics come from a tester, add the spread and the commission on your own terms: on tight stops that changes the conclusion entirely.
check 4Frequently asked questions
Which profit factor counts as good?
For a system with real costs the working range is 1.3–1.8. Higher values require a check of the sample: how many trades, is it one market regime and was the statistic made by a single large trade.
How does the profit factor differ from expectancy?
It is the same information in a different form. Expectancy gives the average result of one trade in R, the profit factor the ratio of the totals. Expectancy is more convenient for distance calculations, the profit factor for a quick comparison.
Can the profit factor of different systems be compared?
Only under comparable conditions: a similar number of trades, one period and the same treatment of costs. Otherwise what is compared is not the systems but the samples.
What to do if the profit factor is below one?
First check whether the costs are the reason: recalculate the result without the spread and the commission. If it is still below one, the system has negative expectancy, and what has to change is the entry or exit rules rather than the position size.
How is the profit factor linked to the win rate?
Through the formula: profit factor = (share of winners × average R) ÷ the share of losers. At a win rate of 45 % and a ratio of 1 : 2 you get 1.64. If the value calculated from the sums differs from the formula, the averages are distorted by outliers.
What is the profit factor of a system with an expectancy of +0.2 R?
It depends on the ratio. At a win rate of 40 % and an R/R of 1 : 2 it is 1.33, at a win rate of 60 % and an R/R of 1 : 1 it is 1.50. The same expectancy gives a different profit factor, which is why the metrics are read together.
Does the swap affect the profit factor?
Yes, it increases the sum of losses and reduces the sum of profits. On swing positions with a negative swap the difference between a calculation with and without the charges can be significant.
Should the profit factor be calculated by month?
It is useful as a robustness check: if the figure stays in a narrow range, the system works steadily; if it jumps from 0.7 to 3.0, individual trades make the result.
Should the profit factor be counted in money or in R?
Either works, but in R it is more correct when trade sizes differ: otherwise a large position gets more weight and the figure describes the distribution of sizes rather than the system.
Which profit factor counts as suspiciously high?
Above 2.5 on a sample smaller than a hundred trades. The check is simple: remove the best trade and recalculate. A fall below 1.2 means there is no stable edge yet.
Read next
