A trading journal produces a handful of numbers, and traders argue endlessly about which one matters. The honest answer is that they only mean something together. A 70% win rate with a profit factor of 0.9 is a losing system. A 35% win rate with an expectancy of +0.4R is a very good one. This article defines each number, shows the arithmetic on one small set of trades, and explains the two mistakes that make each of them lie.
R-multiple: the unit everything else is measured in
Before the trade, your risk is the distance from entry to stop multiplied by your size. Risk $150 on a trade, make $300, and the result is +2R. Lose the full stop and it is −1R. Take a loss at half the stop and it is −0.5R.
Why bother, when money is right there? Three reasons. R makes trades comparable across account sizes and markets: a +1.5R on gold and a +1.5R on NQ are the same quality of outcome. R exposes position-sizing errors: a trade that was supposed to risk 1% and lost 2.3R of the planned risk tells you the stop was moved or the size was wrong. And R is the unit expectancy is best expressed in, because it does not change when you scale the account up.
The requirement is simple: the journal needs your initial stop. Without it there is no R, only money.
Win rate, and why break-even trades must be excluded
Win rate is the number of winning trades divided by the number of closed trades. The detail everyone gets wrong is what counts as a win. A trade that closes +$2 after fees was not a win; it was a scratch. If you count it, a scalper with many tiny positive exits will show a 65% win rate and no money.
The fix is a break-even threshold: results within plus or minus a small amount (say $10 on a $10,000 account, or 0.05R) are classified as break-even and left out of both the numerator and the denominator. SetupBook sets this per account; the demo's main account uses $10.
Average win, average loss and the payoff ratio
Average win is the mean result of the winning trades; average loss the mean of the losers (as a positive number). Their ratio is the payoff ratio, sometimes called the reward-to-risk realised. Average win $180 and average loss $150 gives a payoff of 1.2.
Payoff and win rate are a pair. For any payoff there is a break-even win rate below which you lose money:
break-even win rate = 1 / (1 + payoff)
With a payoff of 1.2 you need to win more than 45.5% of the time. With a payoff of 3 you can be wrong 75% of the time and still break even. Most "which is better, high win rate or big winners" arguments dissolve once you look at the pair.
Profit factor
Profit factor is gross profit divided by gross loss, where gross profit is the sum of all winning trades and gross loss the sum of all losing trades, as a positive number.
| Profit factor | What it usually means |
|---|---|
| below 1.0 | losing money |
| 1.0 to 1.3 | roughly break-even after a bad week |
| 1.3 to 2.0 | a working discretionary system |
| 2.0 to 3.0 | strong, check the sample size |
| above 3.0 | almost always too few trades |
The caveat is sample size. Ten trades with one big winner produce a profit factor of 4 that means nothing. Treat anything under thirty trades as a rumour, not a statistic; SetupBook shows an "n=" badge next to any number computed from fewer than five trades for the same reason.
Expectancy
Expectancy is what one trade earns on average, before you know whether it wins:
expectancy = (win rate × average win) − (loss rate × average loss)
Take the set from earlier: 20 trades, 11 winners averaging $180, 9 losers averaging $150.
| Step | Calculation | Result |
|---|---|---|
| Win rate | 11 / 20 | 0.55 |
| Loss rate | 9 / 20 | 0.45 |
| Expectancy in money | 0.55 × 180 − 0.45 × 150 | +$31.50 per trade |
| Expectancy in R (risk $150) | 31.50 / 150 | +0.21R per trade |
| Profit factor | (11 × 180) / (9 × 150) | 1.47 |
An expectancy of +0.21R means that over a hundred trades you expect to make around 21R. If that sounds small, note that it is 21% of the account per hundred trades at 1% risk, with no leverage tricks. Expectancy is the number to watch month to month; win rate and payoff are how you diagnose why it moved.
Drawdown
Drawdown is the fall from the highest point of your equity curve to the lowest point after it, before a new high is made. Report it in money and as a percent of the peak. An account that went from $10,000 to $11,400, then down to $10,700, is in a $700 drawdown, 6.1% from the peak, even though it is still up on the year.
Maximum drawdown is the largest such fall in the period. It is the number prop firms watch and the number that decides whether you can stick with a strategy emotionally. A system with +0.3R expectancy and a 25% maximum drawdown will be abandoned by most people before it pays.
The two mistakes that make every metric lie
Typing the result instead of deriving it. When the P&L is a number you enter, a sign error on a short, a forgotten fee or a wrong contract multiplier goes straight into every statistic. A journal that computes the result from entry, exit, size and the symbol's multiplier, and warns when a typed result disagrees with the prices, removes most of these at the source.
Mixing fills and trades. A position opened in two fills and closed in three is one trade. If a spreadsheet records it as five rows, the win rate, the average win and the R-multiple are all wrong at once. Keep fills under the trade and let the journal combine them.
Where these numbers live in SetupBook
The dashboard shows net P&L, win rate, profit factor, expectancy, average R, best and worst trade and drawdown for the selected account and period, computed from the fills you recorded. The analytics page breaks each of them down by symbol, strategy, session, setup quality and hour, and the trader score card prints its own formula so you can check it. The glossary in the user guide has every definition in one place, and the free account has no limit on trades, so the sample size problem is the only one you have to solve yourself. Start a journal and the arithmetic is done for you.