What is a good profit factor in a backtest? Range, trades, costs

What is a good profit factor in a backtest?

A good profit factor in a backtest is about 1.3 to 2.0 after costs on a few hundred trades. That answers what is a good profit factor in one line; the longer answer depends on trade count and where the profit came from. Costs mean commission and slippage. Profit factor equals gross profit divided by gross loss.

Gross profit is the sum of all winning trades, and gross loss is the sum of all losing trades. A strategy that won $15 for every $10 it lost has a profit factor of 1.5. A backtest replays a rule on old prices with costs included, so the ratio you read must already contain commission and slippage. Think of the ratio as a balance scale: every dollar lost adds weight to one pan, and every dollar of profit that costs eat away lightens the other pan. Because it is a ratio of two sums, not an average, it has no unit and does not grow with trade count, which makes it easy to compare across strategies and easy to misread. The profit factor formula and its three common pitfalls are covered in detail in their own article.

Is a profit factor of 2 good, and why do very high values mislead?

A profit factor of 2 after costs on a few hundred trades is a strong result, while a value above about 3 in a backtest usually has a cause other than a good rule. The four usual causes are listed in the order you should check them:

1. Too few trades. With 20 trades, two large winners can carry the whole ratio.
2. One or two outsized winners. Remove the three largest winning trades and recompute; a ratio that falls below 1 was never a strategy result, it was a lucky trade.
3. A search that kept the best of many candidates. The winner of a thousand variants carries a selection bonus. Bailey, Borwein, López de Prado and Zhu (2014) show that this bonus grows with the number of variants tried, and the same authors (2017) give a way to estimate how likely the chosen one is overfit. The mechanism is explained in why the best backtest in a search is usually inflated.
4. A leak of future prices. A rule that decides on a bar's close and fills at that same close, or that reads an indicator value not yet available, produces ratios no live account can match.

How many trades do I need to trust a profit factor?

Under about 30 trades a profit factor is a hint, a few hundred trades make it evidence, and the test for any sample is to remove the largest winners and see whether the ratio stays above 1. The reason is the formula: gross profit is a sum, so one large winner can be a third of it. On a small sample that one trade decides the ratio; on a large sample it is one of hundreds.

A real search result shows how to read a sample in the middle. The run below comes from AlphaPrime (Valynox) on NQ daily bars from 2019-01-02 to 2025-06-30, long and short, one contract, with $20 commission and $10 slippage charged per trade. The search tried 1,024 candidates and this is the only one it kept, selected on return against drawdown. The rule it uses is a stand-in built from standard indicators, not a profit factor rule, because profit factor is a measurement and not something you can trade.

The profit factor of 1.636 sits inside the usual band, and the win rate of 54.29% says the ratio comes from winners being larger than losers, not from winning often. But 105 trades is between a hint and evidence, and this candidate was picked after the fact out of 1,024, which flatters it. Nothing here says the rule earned that money or that a reader should expect it; the right next move is the top-winner check and a run on data the search never saw. AlphaPrime prints the profit factor, the trade count and the drawdown on one page of its Performance Report and lists the trades beneath them, so removing the largest winners by hand takes a few minutes.

Does the profit factor survive costs?

Compute the profit factor only after commission and slippage, then rerun the backtest with doubled slippage, because costs hit both pans of the scale at once. Every round-trip cost taken from a winning trade lowers gross profit, and every cost added to a losing trade raises gross loss. The ratio therefore falls faster than the net profit alone suggests, and a strategy that trades often falls fastest.

Three steps make the check complete:

1. Enter commission and slippage before the first run. A ratio read before costs and corrected afterwards hides which trades the costs removed.
2. Run it again with slippage doubled. If the ratio drops below about 1.2, the strategy depends on cheap fills.
3. Read it next to maximum drawdown and the recovery factor. Recovery factor is net profit divided by maximum drawdown, and it tells you what the profit factor cannot: how deep the hole was on the way to the result. Two strategies with a profit factor of 1.6 can have very different drawdowns, which is why the recovery factor and what counts as good belongs on the same page. The free drawdown calculator gives the maximum drawdown from an equity series.

Open your most recent backtest report and write down three numbers: the profit factor after costs, the trade count, and the profit factor again after you remove the three largest winners. If the third number is still above 1 on a few hundred trades, rerun the test with doubled slippage; if it is not, the strategy has not yet shown you anything.