Profit factor is gross profit divided by gross loss. Above 1.0 is profitable. In practice, 1.2 to 1.5 represents a realistic and durable edge on retail timeframes. Figures above 2.5 usually indicate curve fitting, too few trades, or unmodelled costs rather than exceptional skill.
A profit factor of 1.3 means that for every dollar lost, $1.30 was gained. It's a cleaner measure than win rate because it accounts for the size of wins and losses, not just how often you're right. A strategy can win 30% of the time and have an excellent profit factor if the winners are large enough.
Genuine edges in liquid markets are thin, because thousands of participants compete them away. A backtest showing 3.0 or higher on a retail timeframe almost always means the parameters were fitted to the sample, the trade count is small, or costs were ignored. Extraordinary claims require out-of-sample evidence.
Profit factor alone can mislead. Check the trade count (is this 40 trades or 400?), maximum drawdown (can you psychologically survive it?), the worst losing streak, and expectancy per trade net of costs. Our own sweep strategy sits at 1.12 net of spread — modest, honest, and something we'd rather report accurately than inflate.
For a mechanical strategy you intend to trade: above 1.1 net of costs across hundreds of out-of-sample trades is worth taking seriously. Below 1.0 means you're paying the market for the privilege of trading.
Yes — 1.2 net of costs across a large out-of-sample trade count is a genuine, tradeable edge.
Durable systematic strategies commonly sit in the 1.1–1.5 range once costs are included.
In backtests, yes — unusually high figures typically indicate overfitting or too small a sample rather than a superior strategy.