Recovery Factor vs Profit Factor: Read Drawdown Risk

Two trading records can finish with the same profit and the same profit factor, yet take very different routes to the finish. One loses a little between wins. The other collects its losses in a single unpleasant pile. Recovery factor brings the size of that drawdown into the comparison.

Profit factor compares gross winning and losing results. Recovery factor compares net profit with maximum drawdown. Neither is a forecast. For an automated strategy, they describe different parts of the historical record you should understand before connecting signals to execution. In an AlgoWay TradingView automation workflow, alert routing does not turn a favorable test ratio into evidence that future trades will follow the same path.

Keep the two denominators separate

The MetaTrader 5 testing-report reference describes profit factor as gross profit divided by gross loss, and recovery factor as profit divided by maximum drawdown. Use the positive magnitude of gross losses when calculating the familiar positive profit-factor ratio.

• Profit factor = gross profit / absolute gross loss.
• Recovery factor = net profit / maximum drawdown in money.

The second formula needs a declared drawdown basis. Balance and equity are different series. MetaQuotes' MQL5 testing-statistics reference specifically documents its recovery-factor statistic using maximum balance drawdown. The general testing-report help uses the broader wording “maximum drawdown.” When reconciling a displayed ratio or an exported report, check its definition and reproduce the division from the actual fields instead of silently substituting an equity or percentage figure.

A maximum money drawdown is a peak-to-subsequent-trough loss, not the sum of every losing trade. That is why the two ratios can disagree without either being broken.

Same trades, different order, different recovery factor

Consider a simplified account starting at 10,000, with four fixed monetary trade results: +1,000, +1,000, -500 and -500. Assume these are the final net trade results, with no deposits, withdrawals, open positions or additional costs. This is an arithmetic illustration, not a backtest.

Both sequences earn 2,000 in winning trades and lose 1,000 in losing trades. Both therefore have profit factor 2 and net profit 1,000.

Sequence A's largest balance fall is 500. Its balance-based recovery factor is 1,000 / 500 = 2. Sequence B falls from 12,000 to 11,000, so its maximum balance drawdown is 1,000 and its recovery factor is 1,000 / 1,000 = 1.

The rearrangement leaves profit factor unchanged because the winning and losing totals stay unchanged. Recovery factor changes because the path changes. In a real strategy, rearranging trades might also change position sizes and subsequent results. The fixed money amounts here isolate one effect; they do not simulate that feedback.

The ratio is not a recovery clock

A recovery factor of 2 does not mean the account recovered in two days, two trades or twice as fast as another account. There is no time variable in the division. Two equity paths with the same net profit and maximum drawdown could spend very different amounts of time below their previous peaks.

The example also ends below both sequences' highest balances. A positive recovery factor does not establish that the latest drawdown has ended. Inspect the dates of the peak, trough and subsequent recovery separately if time underwater matters to your decision.

Nor should you confuse the ratio with the percentage gain needed to recover a loss. A 20% capital decline needs a 25% gain from the lower balance to return to the starting level. That is a different calculation with a different denominator.

Read the path before ranking the bot

For a useful comparison, retain the test dates, starting capital, sizing method, cost assumptions and drawdown basis beside each ratio. A larger recovery factor from a longer or differently sized test is not automatically a better strategy. Net profit can grow over a longer sample while the largest recorded drawdown remains unchanged; the ratio itself does not normalize for elapsed time.

Check equity drawdown even when the ratio uses balance. Open losses can create an uncomfortable equity path before a position closes. Inspecting only realized results can leave that exposure out of the comparison.

A tiny denominator deserves scrutiny in either metric. Few losing trades can make profit factor unusually sensitive to one additional loss. A small observed drawdown can make recovery factor look impressive without proving that larger drawdowns are unlikely. With a zero denominator, ordinary division is undefined; a software display convention is not proof of risk-free trading.

Keep the full trade sequence and balance/equity curves with the report. Then reserve later data for a separate evaluation and test the execution workflow before relying on live orders. The useful question is not which single number wins the table, but whether you can explain the losses hidden behind it.