Trading metrics boil hundreds of trades down to a handful of numbers. Read well, they tell you whether you have an edge, how much pain to expect along the way and which part of your trading is dead weight. Read badly, they make you abandon a good system or stick with a bad one. This guide explains each metric with its formula, a worked example and the right way to interpret it.
All the examples use the same log of ten closed trades:
| Trade | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
|---|---|---|---|---|---|---|---|---|---|---|
| P&L ($) | +300 | −200 | +150 | +450 | −150 | +200 | −200 | +250 | −100 | +100 |
Six winners totalling $1,450, four losers totalling −$650 and a net result of +$800.
Win rate
The share of winning trades among the trades with a result.
Win rate = winners ÷ (winners + losers) × 100
In the example: 6 ÷ 10 = 60%. Trades that close at breakeven are best left out of the calculation (or always counted the same way): they are neither a win nor a loss, and mixing them in changes the number without any change in your trading.
Win rate alone doesn't tell you whether you make money. An 80% win rate with losses four times bigger than wins will blow up an account; a 35% win rate with wins three times bigger than losses can be an excellent system. Always read it alongside the average size of wins and losses.
Average win, average loss and the win/loss ratio
Win/loss ratio = average win ÷ |average loss|
In the example, the average win is 1,450 ÷ 6 = $241.67 and the average loss 650 ÷ 4 = $162.50, so the ratio is 1.49: each winner makes, on average, almost one and a half times what a loser loses.
Win rate and ratio are tied together. For a given ratio, the minimum win rate to break even (before costs) is:
Breakeven win rate = 1 ÷ (1 + ratio)
| Win/loss ratio | Win rate needed to break even |
|---|---|
| 0.5 | 66.7% |
| 1 | 50% |
| 1.5 | 40% |
| 2 | 33.3% |
| 3 | 25% |
This table is why cutting winners early "to lock it in" is so expensive: it lifts your win rate, but it shrinks your ratio faster than it raises your hit rate.
Profit factor
Profit factor compares everything you won with everything you lost.
Profit factor = gross profit ÷ gross loss
In the example: 1,450 ÷ 650 = 2.23. For every dollar lost, the system made 2.23.
- Below 1: you are losing money.
- Between 1 and 1.3: a very thin edge; commissions and slippage can wipe it out.
- Between 1.3 and 2: a solid edge for most styles.
- Above 2: excellent, but check the sample size before celebrating.
If you have no losing trades at all, profit factor can't be calculated. It isn't "infinite": there just isn't enough data to measure it yet.
Profit factor relates to the two previous metrics like this: PF = (win rate × average win) ÷ ((1 − win rate) × average loss). With a 40% win rate and a ratio of 2, profit factor is 0.4 × 2 ÷ 0.6 = 1.33.
Expectancy
Expectancy is what you make or lose on average per trade. It is the metric that best answers the question "does this system make money?".
Expectancy = (win rate × average win) − (loss rate × average loss) = net P&L ÷ number of trades
In the example: 0.6 × 241.67 − 0.4 × 162.50 = 145 − 65 = $80 per trade, which matches $800 ÷ 10 trades.
A positive, stable expectancy is what you're after. With it, more well-executed trades mean more expected profit; with a negative expectancy, trading more only speeds up the loss.
R-multiples: the unit that makes everything comparable
An R-multiple expresses a trade's result in units of what you risked on it.
R = trade P&L ÷ initial risk
If you risk $200 and make $400, the trade is worth +2R. If you lose the $200, it's −1R. The advantage of R-multiples is that they let you compare trades of different sizes, different accounts and different periods.
| Trade | Risk | P&L | Result in R |
|---|---|---|---|
| 1 | $200 | +$400 | +2R |
| 2 | $200 | −$200 | −1R |
| 3 | $200 | +$100 | +0.5R |
| 4 | $200 | −$200 | −1R |
| 5 | $200 | +$600 | +3R |
Total: +3.5R over five trades, an expectancy of +0.7R per trade. A loss bigger than −1R is a red flag: it means the stop wasn't respected or the size was bigger than planned.
Max drawdown
Drawdown is the fall from the highest point of your equity curve to the lowest point after it. Max drawdown is the biggest of those falls in the period.
Drawdown = (peak − trough) ÷ peak × 100
If your account goes from $10,000 to a high of $12,000 and then falls to $10,200, the drawdown is $1,800, or 15% of the peak. It measures the pain the system has had to endure, and it's the metric that decides whether you'll be able to keep trading it when the bad streaks come.
Losses aren't symmetrical: the deeper the hole, the harder it is to climb out.
Gain needed to recover = loss ÷ (1 − loss)
| Account drop | Gain needed to recover |
|---|---|
| −10% | +11.1% |
| −20% | +25% |
| −30% | +42.9% |
| −40% | +66.7% |
| −50% | +100% |
That's why risk management matters more than accuracy: avoiding a 40% drawdown is worth more than almost any improvement to your entries.
Recovery factor
Recovery factor = net profit ÷ |max drawdown|
If a system has made $5,400 with a max drawdown of $1,800, its recovery factor is 3: it has earned three times what it lost in its worst stretch. Below 1, the system hasn't yet made up for its worst fall.
Losing streaks: how many are normal
Losing streaks don't necessarily mean something is wrong. They are a mathematical consequence of your win rate. As a rough guide, this is the longest losing streak to expect over 100 trades:
| Win rate | Typical longest losing streak in 100 trades |
|---|---|
| 30% | about 11 losses in a row |
| 40% | about 8 |
| 50% | about 6 |
| 60% | about 5 |
These are averages: in many samples the streak will be longer. If you risk 2% per trade, eight losses in a row are 15% of the account. Knowing this number in advance stops you from switching systems in the middle of a perfectly normal streak.
Consistency
Two traders can make the same in a month: one with twenty similar days and the other with one huge day. The second depends on luck far more than it looks. A simple way to measure it is the weight of the best day:
Best-day share = best day's profit ÷ sum of winning days × 100
If your winning days add up to $5,000 and the best one contributes $2,000, the share is 40%. Many prop firms cap this percentage with their consistency rule. The lower it is, the more spread out and repeatable your result.
How many trades you need before trusting your metrics
With few trades, chance dominates. The standard error of your win rate can be estimated like this:
Standard error = √(p × (1 − p) ÷ n)
With a true win rate of 50%, after 30 trades the standard error is about 9 points: your measured win rate could easily land anywhere between 32% and 68%. With 100 trades the margin narrows to about ±10 points, and with 300 to about ±6. Rule of thumb: don't draw conclusions about a setup with fewer than 30 trades, and don't overhaul your whole system with fewer than 100.
How to read the metrics together
| What you see | What it usually means |
|---|---|
| High win rate, low ratio | You cut winners early or let losers run. Review your manual exits. |
| Low win rate, high ratio | Trend-following style: normal, but prepare for long losing streaks. |
| High profit factor on few trades | You don't know anything yet. Keep building the sample. |
| Positive expectancy, huge drawdown | Risk per trade is too high for the system. |
| Losses bigger than −1R | Moved stops or bigger size than planned: a discipline problem, not a system problem. |
| One day explains the month | An inconsistent result. Look at what happened that day and whether it's repeatable. |
NactionX calculates all these metrics automatically from your journal, filters them by account, setup, hour or session and sums them up in the NX Score, a 0 to 100 score combining profit factor, win/loss ratio, drawdown, win rate, recovery and consistency. If you don't keep records yet, start with the guide on how to keep a trading journal.