Backtestify
Sign in
All articles
Profit Factor ExplainedOctober 1, 202611 min read

Profit Factor: Why 1.3–2.0 Alone Misleads Traders on After Costs

Learn an execution first view of profit factor: compute it after costs, check trade count and equity curve, and verify out of sample results before...

!Isometric illustration of profit factor validation

Profit factor is total gross profit divided by absolute total gross loss, and a value above 1 means a strategy made more than it lost over the test period. Anything from 1.3 to 1.5 is modestly healthy, while readings near 2.0 are stronger, but the number only means something once you check it against drawdown, trade count and costs.

TL;DR:

  • A profit factor below 1 indicates a losing strategy, while values between 1.3 and 1.5 are modestly positive but require careful risk management.
  • A profit factor of 2.0 is strong but must be verified against drawdown, trade count, and out-of-sample performance to confirm robustness.
  • Calculating profit factor after realistic trading costs and stress testing assumptions is essential for trustworthy backtest results.
  • Large profit factor figures can be misleading if driven by one or two outsized trades, making additional metrics like drawdown and win rate crucial.
  • Out-of-sample profit factor typically decreases once costs are included, emphasizing the need for realistic testing and caution about in-sample overfitting.

Table of Contents

2. What is profit factor and how is it defined?

Profit factor compares everything a strategy earned on winning trades against everything it gave back on losing trades, using the same currency and the same test window for both sides of the ratio. The standard version, sometimes called total profit factor, sums every winning trade's profit and loss and divides it by the absolute value of every losing trade's profit and loss combined. A less common variant, average profit factor, divides the average winning trade by the average losing trade instead of the totals, which produces a different number from the same trade list and should never be mixed with the total version in a comparison.

The input numbers matter as much as the formula itself.

  • Total PF: sum of winning trade P&L divided by the absolute sum of losing trade P&L
  • Average PF: average win divided by average loss, used less often and not interchangeable with total PF
  • Both versions should use profit and loss figures calculated after realized trading costs when you are comparing strategies against each other

3. How do you calculate profit factor step by step?

Take a strategy with ten trades: six winners totaling $1,200 and four losers totaling $600. Divide $1,200 by $600 and the profit factor is 2.0, meaning the strategy produced two dollars of gross profit for every dollar it lost. A list of trades you can build in a spreadsheet looks like this:

  1. List every trade's net profit or loss in one column, after costs.
  2. Sum all positive values to get gross profit.
  3. Sum the absolute value of all negative values to get gross loss.
  4. Divide gross profit by gross loss to get profit factor.

Two edge cases break the formula. A strategy with no losing trades has an undefined profit factor since you cannot divide by zero, and a single oversized winner can inflate PF even when most trades lose money, so the ratio alone hides a fragile strategy.

4. What counts as a good profit factor in practice?

A profit factor below 1 means the strategy is losing money outright, and exactly 1 is a break-even system before considering costs that were not included in the calculation. From there, the ranges traders commonly use are rough guides rather than fixed rules:

  • Below 1: losing strategy, stop here regardless of other metrics
  • 1 to 1.3: marginal, usually not worth the operational risk
  • 1.3 to 1.5: modestly positive, workable with tight risk control
  • 1.75 to 2.0 and above: often considered strong, though it deserves a closer look rather than automatic trust

What counts as good shifts with context. A scalping strategy with thousands of trades and a PF of 1.3 can be more statistically reliable than a swing strategy with forty trades and a PF of 2.5, because sample size changes how much confidence you can place in either number. PF also needs to be weighed against net return and maximum drawdown: a high PF on a strategy that barely trades, or one that occasionally suffers a severe losing streak, is not automatically the better choice.

5. Where profit factor falls short and what to check alongside it

Profit factor is a single ratio, and ratios hide the shape of the data that produced them. Academic and practitioner evaluations warn that a high profit factor can come from one outsized winning trade, which means the metric says nothing about trade sequencing, losing streaks, drawdown depth, volatility, leverage or how concentrated the gains are in a handful of trades.

Before trusting a PF number, review it next to these:

  • Net return: the actual dollar or percentage outcome, since a high PF on tiny position sizes may not be worth trading
  • Maximum drawdown: how far the equity curve fell before recovering
  • Expectancy: the average amount won or lost per trade
  • Win rate: the share of trades that were profitable
  • Average win and average loss: whether gains come from many small wins or a few large ones
  • Trade count: how many data points support the ratio
  • Equity curve shape: whether gains are steady or concentrated in a short stretch

Pro Tip: Sort trades by profit and loss and check what share of total gross profit comes from the top three winners. If removing them flips the strategy unprofitable, the profit factor is not telling you what you think it is.

6. How to use profit factor correctly in a backtest

A profit factor calculated before costs is close to meaningless for decision making, since commissions, spreads, slippage and financing can turn an apparently profitable system into a losing one once applied consistently across every trade. The Duke backtesting protocol treats this as a primary source of false confidence in backtested results.

A workable sequence:

  1. Recalculate every trade's profit and loss after realistic commissions, spreads, slippage and financing, then recompute PF on that adjusted figure.
  2. Lock the rule set before testing, reserve a period of data the rules never touched, and report PF separately for the in-sample and out-of-sample periods.
  3. Stress-test the assumptions: widen slippage, raise costs, randomize trade order with a Monte Carlo approach, and run walk-forward validation across rolling windows.
  4. Record how many rule variants were tried before settling on the final version, since testing many combinations on the same data raises the odds that the best in-sample result is a fluke.

A 2026 preprint on the GT-Score found that building anti-overfitting structure directly into the optimization objective improved the generalization ratio by about 98% versus a baseline objective in walk-forward tests on S&P 50 data from 2010 to 2024. The point for most traders without that infrastructure is simpler: an in-sample profit factor that falls apart out-of-sample was never a real edge to begin with.

7. Where profit factor shows up and what a credible report includes

Profit factor appears in exchange and broker trade analytics, manually built spreadsheets, and dedicated backtesting platforms, each with different levels of rigor behind the number. A report worth trusting states the profit factor after costs, the trade count behind it, whether the figure is in-sample or out-of-sample, and shows an equity curve with the drawdown visible rather than just the headline ratio.

  • After-cost PF, not a raw gross figure
  • Trade count and test period stated explicitly
  • In-sample and out-of-sample PF reported separately
  • Equity curve and drawdown chart included alongside the number

Backtestify publishes a library of tested strategies with both in-sample and out-of-sample profit factor shown side by side, built around realistic execution costs rather than frictionless assumptions.

8. What to take away about profit factor

Profit factor is useful as a quick screen, not a verdict: a number above roughly 1.3 to 1.5 is worth a second look, never a green light on its own. Before acting on any PF figure, compute it after costs, check the trade count and equity curve, and favor out-of-sample results over whatever looked best during optimization.

!Four checks for validating profit factor

9. What testing published strategies reveals about profit factor

Across strategies built from popular trading content, profit factor tends to erode once realistic spreads, commissions and slippage replace the frictionless assumptions many creators use when presenting results. A strategy that looks strong in-sample often loses a meaningful chunk of its edge out-of-sample, which is why out-of-sample PF deserves more weight than whatever number appeared during the optimization run. The pattern shows up often enough that it should change how any trader reads a PF claim attached to a strategy they did not test themselves.

— WAJDI

10. Verify your own profit factor before risking capital

Running the backtesting protocol described above by hand, across costs, out-of-sample splits and stress tests, takes real spreadsheet work for every strategy you want to check. Backtestify runs that process for you: enter a strategy's rules in plain English and get a report with profit factor, win rate and maximum drawdown shown separately for in-sample and out-of-sample periods, after realistic execution costs.

Backtestify

The platform's published strategy library includes tested versions of well-known approaches, including a moving-average pullback strategy and a sweep and fair value gap setup, so you can see the report format before testing your own rules. The Free plan lets you try the core workflow, and the Pro plan unlocks unlimited backtesting, improvement and forecasting. Current prices are available on the pricing page. Check current plans or see the methodology behind how trades are simulated.

11. Research behind backtesting and overfitting risk

Sources

FAQ

Is a 2.0 profit factor good?

A profit factor of 2.0 is generally considered strong, since it means a strategy earned roughly twice as much on winning trades as it lost on losing ones. It still needs to be checked against trade count, drawdown and whether the figure holds up on out-of-sample data before it means much.

Is a profit factor of 1.5 good?

A profit factor of 1.5 is modestly positive and workable for many strategies, especially with a reasonable trade count and controlled drawdown. It sits below the stronger 1.75 to 2.0 range but well above the break-even point of 1.

Is 1.3 a good profit factor?

A profit factor of 1.3 is on the lower edge of acceptable, often only worth trading when paired with tight risk control and a large enough sample size to trust the number. On a small trade count it carries more uncertainty than a 1.3 on a strategy with hundreds of trades behind it.

How is profit factor calculated?

Profit factor is calculated by summing the profit from every winning trade, summing the absolute loss from every losing trade, then dividing the first number by the second. A strategy with $1,200 in gross profit and $600 in gross loss has a profit factor of 2.0.

Recommended

Read next

8 Step TradingView Pine Script Backtest to Catch Repaint and Fill Bugs

Continue

Want these checks applied to your own rules automatically?

Run a backtest