Win rate, profit factor and payoff ratio: reading trade statistics

On this page

Key takeaways

  • Win rate counts how often trades end with a gain, not how much they gain. On its own it says nothing about profit.
  • The payoff ratio compares the size of the average win with the size of the average loss. Trend-following rules usually win less often with large payoffs; dip-buying rules win often with small ones.
  • Expectancy, the average result per trade, combines the two. It is what decides whether a rule made money before costs, and costs come out of it on every trade.
  • Profit factor divides everything gained by everything lost. MarketHeist's results table measures it day by day over the strategy's returns, not trade by trade, so it is not the trade-level figure many platforms quote.
  • Trade counts matter. With a few dozen trades a win rate or payoff ratio can swing a long way by luck, and a handful of large trades can decide the whole result.

Definition

Win rate is the share of a strategy's closed trades that ended with a gain; the payoff ratio is the average winning trade divided by the size of the average losing trade; and profit factor is total gains divided by total losses. None of them alone says whether a strategy made money: a high win rate with small wins and large losses can lose, and a low win rate with large wins can earn, which is why they are read together as expectancy, the average result per trade.

Also known as: win rate, hit rate, percent profitable, profit factor, payoff ratio, average win to average loss, expectancy, risk-reward ratio, trade statistics.

How it works

A backtest's trade list records each round trip: when the strategy went in, when it came out, and what it made or lost in between. Trade statistics summarise that list. The win rate is the share of trades that ended with a gain. The average win and the average loss are the mean returns of the winning and the losing trades. The payoff ratio divides the first by the size of the second, so a payoff of two means the typical winner made twice what the typical loser lost.

Win rate and payoff are two halves of one quantity. Expectancy, the average return per trade, is the win rate times the average win plus the loss rate times the average loss. A rule that wins most of the time but loses much more on its losing trades than it gains on its winners can have negative expectancy; a rule that wins rarely but whose winners dwarf its losers can have positive expectancy. Each payoff ratio has a breakeven win rate, shown in the formula below, and the larger the payoff, the lower the win rate a rule needs.

Trade statistics are like a shop's sales figures: the share of visitors who buy says little until you know how much each sale earns and what each refund costs.

Profit factor is total gains over total losses; above one, the gains won. Trade by trade, it equals the payoff ratio times the number of winners per loser. MarketHeist's results table computes it on the strategy's daily returns instead: the sum of positive daily returns over the sum of negative ones, with days in cash contributing nothing. That makes it a property of the return stream, comparable between rules whose trades last a day or a year, but it is not the number a trade list gives, and the two can disagree.

The mix of win rate and payoff comes from a rule's design. Trend-following rules exit quickly when a trend fails and hold winners for months, so they lose often and win big. Dip-buying rules buy a fall and sell the bounce, so they win often and small, and their occasional large loss comes when the fall keeps going. Neither profile is superior on its face; what matters is expectancy after costs, how long the worst run of losses lasts, and how few trades the statistics rest on.

Formula

Win rate, payoff ratio and expectancy
W=n+n,R=a¯+|a¯−|,E=Wa¯++(1−W)a¯−
n
number of closed trades
n₊
number of trades with a return above zero
W
win rate
ā₊
average return of the winning trades
ā₋
average return of the losing trades (negative)
R
payoff ratio
E
expectancy: the average return per trade
Breakeven win rate for a given payoff ratio
E>0⇔W>W*=11+R
W*
the win rate at which expectancy is zero

With wins as large as losses (R = 1) a rule needs to win more than half its trades; with wins four times its losses (R = 4), more than one in five.

Profit factor: per trade, and per bar as MarketHeist reports it
PFtrade=∑a+∑|a−|=RW1−W,PFbar=∑tmax(rt,0)∑tmax(−rt,0)
a₊, a₋
returns of the individual winning and losing trades
rₜ
the strategy's return on bar t, after costs (0 on days in cash, minus the cost on an exit day)
PF_trade
profit factor over trades, as a trade list reports it
PF_bar
profit factor over bars, as MarketHeist's results table reports it

Both are above 1 when gains exceed losses. They differ because a winning trade contains losing days and a losing trade winning ones.

Worked example

SPY's daily bars from Oct 2006 to Sep 2026, about 20 years, with two rules of opposite design, both acted on with the standard one-bar delay, no trading costs and no interest on cash: a trend rule that holds SPY while its close is above its 200-day simple moving average, and a dip-buying rule that holds SPY only while its 2-day RSI is below 10, a deeply oversold reading.

The trend rule completed 56 round trips. Only 25.0% of them ended with a gain, but the average winner made 18.2% against an average loss of -1.5%, a payoff ratio of 12.43. Its trades lasted 69.1 trading days on average: many short exits after false breaks of the average, and a few long holds through rising markets.

The dip-buying rule traded far more often, 265 round trips lasting 1.8 days on average, and won more often: 78.1% of its trades ended with a gain. Its wins were small, 1.1% on average against an average loss of -1.4%, a payoff ratio of 0.75.

Expectancy brings the two together. Per trade, the trend rule averaged 3.5% and the dip-buying rule 0.5%: the rule that lost most of its trades earned more on each one, because its few winners were large. For the trend rule, its win rate times its average win plus its loss rate times its average loss gives back the 3.5% average.

The results table's profit factor, measured over daily returns rather than trades, was 1.17 for the trend rule and 1.70 for the dip-buying rule. Over the whole window the trend rule grew 9.1% a year and the dip-buying rule 7.1%, against 11.1% for holding SPY. With 56 trades, the trend rule's statistics rest on a handful of long holds; the dip-buying rule's 265 trades give a steadier estimate of its win rate, though not of the rare fall that keeps going.

Data through Sep 2026. Reproduce id 8e25813a4747a651 (this guide, its lab setup and the data month).

Interactive lab

Both rules' trades marked on SPY's price, with win rate, average win and loss, payoff ratio, expectancy, trades and the per-bar profit factor beside buy-and-hold. Change either rule's settings, add a cost, or switch funds.

SPY price with the days the base rule held itS&P 500, adjusted daily close, Oct 2006 to Sep 2026, log scale. Shaded: the 57 stretches the base rule held the fund (79% of 5030 days). Also drawn: 200-day simple average.$50$100$200$50020062011201620212026SPY adjusted close200-day simple averageHeld by the base rule
SPY on a log scale. Shaded: the days the base rule held the fund. Oct 2006 to Sep 2026.
Short / long average ratio with the base rule's lineShort / long average ratio on SPY, Oct 2006 to Sep 2026. The rule compares it with the fixed line 1.0.70.80.911.11.2200620112016202120261Short / long average ratio
Short / long average ratio and the line the base rule compares it with. Oct 2006 to Sep 2026.
Growth of $10,000: the setups against buy and holdS&P 500, Oct 2006 to Sep 2026, log scale. $10,000 grows to $56,592 (base), $39,526 (Dip-buying: RSI(2) below 10), $82,292 (buy and hold).$10k$20k$50k20062011201620212026Base ruleDip-buying: RSI(2) below 10Buy and hold
Growth of $10,000 on a log scale, same days for every line. Oct 2006 to Sep 2026.
MeasureBaseDip-buying: RSI(2) below 10Buy and hold
CAGR9.1%7.1%11.1%
Volatility (annualized)11.7%9.3%19.4%
Maximum drawdown-20.7%-15.5%-55.2%
Sharpe ratio0.800.790.64
Sortino ratio1.091.46n/a
Growth of $10,000$56,592$39,526$82,292
Time in the market79.3%9.5%100.0%
Round trips56265n/a
Round trips per year2.813.3n/a
Win rate25.0%78.1%n/a
Average win18.2%1.1%n/a
Average loss-1.5%-1.4%n/a
Payoff ratio12.430.75n/a
Average days held69.11.8n/a
Expectancy per trade3.5%0.5%n/a
Profit factor, per day1.171.701.13
Profit factor, per trade4.142.69n/a
Indicator, last day1.06513.902n/a
Indicator, 10th / 50th / 90th percentile0.944 / 1.063 / 1.12110.439 / 63.38 / 95.153n/a
Days the rule alone is long79.3%9.5%n/a

S&P 500, Oct 2006 to Sep 2026, 5030 trading days. Every setup is measured over the same days, with the indicator warmed up on earlier ones; cash earns 0. Base: MA Crossover (short window 1, long window 200, ma type SMA); long while the Short / long average ratio is above 1; execution delay 1; cost 0 bps one way. Dip-buying: RSI(2) below 10: RSI (period 2); long while the RSI is below 10; execution delay 1; cost 0 bps one way.

Buy and hold, same daysValue
Largest one-day gain14.5% on 13 Oct 2008
Largest one-day loss-10.9% on 16 Mar 2020
CAGR without the 10 largest gains6.6%
CAGR without the 30 largest gains1.8%
CAGR without the 10 largest losses16.0%
CAGR without the 30 largest losses22.1%

The days replaced by 0 are the largest daily gains (or losses) of the window.

How MarketHeist computes it

Every choice an implementation has to make, as the code makes it today. The page is checked against the code each time it is built.

ChoiceHow it is computedIn the code
DataDaily bars (open, high, low, close, volume) of the fund as the MarketHeist data service serves them to the backtester: prices adjusted for splits and dividends, so a dividend counts as reinvested. They come from a snapshot taken each month and cut at the data month of this page, and are not rounded. The fund's own history, not an extension: a fund younger than the window is measured over the years it has.
WindowThe worked example and the lab measure the last 20 years of trading days, Oct 2006 to Sep 2026 (5030 days), ending at the data month of this page. Every setup, and buy and hold, is measured over exactly these days; an indicator is warmed up on the days before them, so changing a setting never moves the window, and a setting that needs more history than exists before the window is refused rather than measured over fewer days.
BacktestThe backtester's own engine, in the order it runs it: the indicator, the position rule (or entry and exit triggers) and the optional regime filter produce a position for every day of the fund's whole history; the position is then shifted forward by the execution delay and applied to each day's return over the window. Cash earns 0, there is no leverage, and a one-way cost in basis points is charged on the size of every change of position. Round trips are the closed entries of the trade log (a trade still open on the last day is not counted); the win rate is the share of them with a gain; the payoff ratio is the average win over the size of the average loss; expectancy is the average return per trade; the per-trade profit factor is the sum of the winning trades' returns over the size of the sum of the losing ones, while the per-day profit factor sums days.
Years and volatilityA year is 252 trading days, the engine's convention: trades per year is round trips over days divided by 252, the CAGR is the growth multiple to the power 252 over the days, and volatility is the sample standard deviation of the daily returns times the square root of 252.
Indicator statisticsThe indicator's main output over the window: its last value, its 10th, 50th and 90th percentile (linear interpolation between ranks), and the share of days the rule alone says long, before the delay and any regime filter. Every output of the indicator is also summarized by its median and the share of days below 0.
Buy and holdThe fund held every day of the window with dividends reinvested. The largest one-day gain and loss are the highest and lowest daily returns; the CAGR without the 10 or 30 largest gains (or losses) replaces those days' returns with 0 and recompounds.
What a trade isOne stretch of consecutive invested bars after the execution delay: the entry is the first invested bar, the exit the last. A trade still open on the last bar of the window is shown in the trade log but not counted among closed trades.computeTradeLogpackages/shared/engine/backtest.ts
Trade returnStrategy equity on the last invested bar ÷ equity on the bar before entry − 1, after costs and leverage. It includes the entry cost; the exit cost falls on the first bar in cash and so is in the growth figures but not in the trade's return.computeTradeLogpackages/shared/engine/backtest.ts
Win rate and averagesA winning trade has a return above zero. Average win, average loss and expectancy are simple means of trade returns, not weighted by how long a trade lasted. Payoff = average win ÷ |average loss|. Bars held counts invested bars per trade.
Profit factor in the results tableSum of positive bar returns ÷ |sum of negative bar returns| over every bar of the strategy, including cost-only exit bars; days in cash return exactly 0 and count on neither side. Infinite when no bar lost. Buy-and-hold's profit factor uses its own daily returns.profitFactorpackages/shared/engine/metrics.ts
Engine functionsThe functions the portfolio app and the MarketHeist MCP tools run, called here on the same monthly returns.computePositionpackages/shared/engine/run_basic.tsgeneratePositionpackages/shared/engine/positions.tsapplyRegimeFilterpackages/shared/engine/regime_filter.tsgenerateTriggeredPositionpackages/shared/engine/triggered_position.tsrunBacktestpackages/shared/engine/backtest.tsgetIndicatorpackages/shared/engine/indicators/registry.ts

Win rate, profit factor and trade statistics compared with its neighbours

Trade statistics and what each one misses
StatisticWhat it measuresBlind spot
Win rateShare of trades with a gainSays nothing about the size of wins and losses
Payoff ratioAverage win over average lossSays nothing about how often each happens
ExpectancyAverage return per tradeIgnores how long trades last and how bumpy the ride was
Profit factor (per trade)Total gains over total losses across tradesDominated by a few large trades when there are few trades
Profit factor (per bar, MarketHeist)Gains over losses across daily returnsNot comparable with a trade-list figure
Sharpe ratioReturn per unit of volatilityTreats large gains as risk; ignores trade structure
Maximum drawdownWorst fall from a peakOne episode; says nothing about how often it happens

Limitations

  • Few trades, wide error. A slow rule may complete only a few dozen trades in two decades, and its win rate and payoff can change a lot if a handful of trades go the other way.
  • Statistics before costs. Costs reduce every trade's result, and they hit dip-buying rules hardest, because their wins are small and their trades many.
  • Averages hide the order of trades. A run of losses in a row is what traders abandon rules over, and neither win rate nor expectancy shows how long such runs were.
  • A trade's return excludes its exit cost, which lands on the next bar, so with costs set the trade statistics read slightly more favourably than the growth figures.
  • One fund, one window. A dip-buying rule's high win rate on a stock index that rose for most of the window says little about an asset in a long decline.

Frequently asked questions

What is a good win rate for a trading strategy?
There is no target figure on its own. What matters is win rate together with payoff: a rule that wins one trade in three can make money if its winners are several times its losers, and a rule that wins most of its trades can lose if its few losers are large. Check expectancy, and then whether it survives costs.
What is the difference between profit factor and payoff ratio?
The payoff ratio compares the average winner with the average loser. Profit factor compares total gains with total losses, so it also reflects how many winners and losers there were: per trade, it equals the payoff ratio times the number of winners per loser.
Why does MarketHeist's profit factor differ from my trade list?
MarketHeist's results table computes profit factor over daily returns: all positive days against all negative days while invested. A trade list computes it over whole trades. A winning trade usually contains many losing days, so the two numbers measure different things and can differ widely.
Can a strategy with a high win rate lose money?
Yes. If losing trades are much larger than winning ones, the losses outweigh the frequent small gains. Dip-buying rules often have this profile: many small wins and an occasional large loss when a fall keeps going.
How many trades are enough to trust a win rate?
More than most backtests have. The uncertainty of a win rate shrinks only with the square root of the number of trades, so a few dozen trades leave a wide range of plausible true values. Compare the result across nearby settings and other funds before reading much into it.

References

  1. Tharp, V. K. (1998) Trade Your Way to Financial Freedom. McGraw-Hill. Expectancy as win rate and payoff combined.
  2. Vince, R. (1990) Portfolio Management Formulas. Wiley. Trade statistics and the mathematics of position sizing.
  3. Kaufman, P. J. (2013) Trading Systems and Methods. Wiley, fifth edition. Trend-following and mean-reversion trade profiles and their statistics.

Try it on your own holdings

Open the backtester and read the trade log →

Or ask an assistant connected to the MarketHeist MCP server (https://marketheist.io/mcp), which computes the same figures:

Using MarketHeist.io, backtest two rules on SPY daily bars with execution delay 1: hold while the close is above its 200-day simple moving average, and hold while the 2-day RSI is below 10. Compare their number of trades, CAGR and profit factor with buy-and-hold, and explain why a rule that loses most of its trades can still earn more per trade.

Ask an AI assistant

The prompt opens in a new tab, already filled in. An assistant without the MarketHeist connector answers from its own knowledge, not from the figures on this page.

Historical simulation, not advice. Computed by the MarketHeist portfolio engine from monthly total returns, data through Sep 2026. Markdown version of this page: win-rate-and-profit-factor.md.