One number for whether the edge is real
Profit factor cuts through win rate and trade count: over a dollar won per dollar lost, or not. Pair it with sample size and the expectancy calculator before trusting a backtest.
One number for "does this work"
Profit factor is gross profits divided by gross losses. Win $6,000 total, lose $4,000 total: PF = 1.5. Below 1.0 the strategy loses money, full stop, regardless of win rate or how good the entries feel.
Interpretation bands from real trading: 1.0–1.2 is noise — one bad week erases it and fees probably already did. 1.3–1.75 is a real edge worth scaling carefully. Above 2.0 on a decent sample, verify before celebrating — most PF > 2 results come from small samples or one outlier trade carrying everything. Recompute with your best trade excluded; if PF collapses below 1.2, you don't have a strategy, you have a memory.
PF also pairs with win rate to describe your style: scalpers run high win rates with PF near 1.3; trend traders run 35% win rates with PF near 1.8. Neither is better — but knowing which you are stops you judging one by the other's numbers.