Profit Factor

What is the profit factor?

Profit factor is the ratio of a strategy's gross profits to its gross losses over a test period: the sum of all winning trades divided by the absolute sum of all losing trades. A value above 1.0 means the system made money; 1.5 means winners collectively earned one and a half times what losers cost; below 1.0 the strategy lost money before any allowance for the capital tied up earning it.

The metric earns its ubiquity in backtesting reports by being dimensionless and immediately interpretable. It does not depend on account size, position sizing, or trade count in the way currency totals do, so it allows quick comparison across markets and configurations. It is also usefully complementary to win rate: a system can win rarely and still carry a strong profit factor if winners are large, or win constantly and sit near 1.0 because losses are outsized.

Its blind spots are equally well known. Profit factor says nothing about the sequencing of results, so it cannot distinguish a smooth equity curve from one with brutal losing streaks, and it is sensitive to outliers: a single enormous winner can prop up an otherwise mediocre trade list. Practitioners therefore read it alongside expectancy, drawdown statistics, and trade counts rather than in isolation.

How it's calculated

Profit factor is a simple ratio of aggregate wins to aggregate losses.

Undefined when there are no losing trades; most software reports it as infinite, which in a backtest is a red flag about sample size rather than a virtue.

Always compute net of commissions and slippage; gross-of-cost profit factors materially overstate quality, especially for high-frequency systems.

How traders use it

• As a first-pass screen on backtests: many practitioners treat net profit factors around 1.5 to 2.0 on a large trade sample as promising, and view values far above 3 with suspicion of overfitting or tiny samples rather than delight.
• To monitor robustness under stress, recomputing profit factor across cost sensitivity sweeps and out-of-sample segments and watching how quickly it decays toward 1.0.
• As a sanity check on outlier dependence: recomputing after removing the single best trade shows whether the edge is broad or hostage to one lucky fill.
• Per-regime or per-segment, splitting the trade list by year, session, or volatility state to see whether profitability is distributed or concentrated in one era.
• Not as a proxy for risk-adjusted return; a strategy can post a fine profit factor while spending months in drawdown, which return/risk ratios like the Sharpe ratio are built to expose.

Profit factor vs related metrics