Win rate is the percentage of trades closed in profit out of all trades taken. It is the most quoted statistic in trading and the most misused, because it counts trades without ever weighing them.
Out of 78 trades
Contents
Win rate measures a frequency: how many times out of a hundred you close a position in the green. A +$3 trade and a +$3,000 trade each count as one, and that is exactly what makes the figure misleading on its own.
It therefore says nothing about profitability. A system can win eight times out of ten and drain an account, if the two losses are each worth six wins. Conversely, a system losing two times out of three can be highly profitable, and that is the case for most trend-following methods.
Win rate = winning trades ÷ total trades × 100
The break-even threshold is what actually matters, and it is computable: minimum win rate = 1 ÷ (1 + RRR).
With a win/loss ratio of 2 you therefore need to pass 33% to break even; with a ratio of 1 you need to pass 50%; with a ratio of 0.5 you need to pass 67%.
Trades closed exactly at break-even are neither winners nor losers: counting them as wins inflates the figure without earning anything.
| Your RRR | Minimum win rate to break even |
|---|---|
| 0.5 | 67% |
| 1 | 50% |
| 1.5 | 40% |
| 2 | 33% |
| 3 | 25% |
| 5 | 17% |
That table is the whole reading of win rate: there is no good figure in the absolute, there is a figure that is sufficient for your ratio. Place your win rate and your RRR on this grid and you know immediately whether your system wins, before even looking at your P&L.
These thresholds are break-even points, excluding fees. In practice you need to clear them comfortably to cover commissions and spread, all the more so if you trade often.
38.5% against a 31.2% threshold: the system is profitable, while losing six times out of ten. Without the RRR, that 38.5% would have been read as a failure.
Place your win rate and your win/loss ratio on the threshold table: if your win rate clears 1 ÷ (1 + RRR), the system wins before fees. That grid is the whole risk to return win rate relationship in one table, and the check takes ten seconds and depends on no lucky period.
In this page's example, 38.5% against a 31.2% threshold: the system is profitable while losing six times out of ten. That is exactly the case win rate alone makes look like a failure.
Win rate does not say whether you win, it says how often you will be wrong. A 35% method means losing two trades out of three for years, and many traders abandon it for that reason alone, although it works.
Knowing that figure before choosing a method avoids adopting one you will not hold.
The first flaw is how break-even trades are treated. Depending on the tool, they are counted as winners, kept in the denominator or removed from the calculation, and the same set of trades shows several different win rates.
The second concerns partial exits. A trade where part was banked before the rest was closed at break-even is often counted as a winner. Win rate rises, the result does not move.
The third is win rate shown on its own, as in many signal offers. Without the win/loss ratio, 85% says nothing, and it very often comes with rare but huge losses.
The classic mistake. Pushing it up by cutting winners early and letting losers run. It is the most natural move in the world, it improves the displayed figure every time, and it destroys the account. A win rate rising while P&L falls is an alarm, not progress.
| Indicator | Formula | What it measures | When to use it |
|---|---|---|---|
| Win rate | winners ÷ non-neutral trades | How often you win | Understanding the psychological comfort |
| RRR | average win ÷ average loss | How winners weigh against losers | Setting targets and stops |
| Profit factor | Σ gains ÷ |Σ losses| | What you earn per $1 lost | Judging a whole system |
| Expectancy | net P&L ÷ number of trades | What an average trade earns | Estimating what more volume will earn |
Moving targets closer raises win rate every time and lowers RRR just as much. The only useful goal is the gap between your win rate and 1 ÷ (1 + RRR), not the figure itself.
A global win rate mixes configurations that have nothing in common. Per setup, it shows which one makes you lose often, information the average erases.
Over twenty trades, three results either way move win rate by fifteen points. Below a hundred, the figure mostly describes the luck of the period.
No figure is good in itself. A 35% win rate is excellent with a win/loss ratio of 3, and disastrous with a ratio of 0.5. The only valid benchmark is the 1 ÷ (1 + RRR) threshold, which depends on how you exit.
Divide the number of trades closed in profit by the total number of trades, then multiply by a hundred. Trades exited exactly at break-even are removed from the calculation, otherwise they inflate the rate without having earned anything. Our win rate calculator does it and shows the matching break-even threshold.
Yes, and it is the common profile of trend-following methods. At 30%, you need a win/loss ratio above 2.33 to break even. Many systems get there because they let a few big moves run and cut the rest quickly.
Profit factor, if you only keep one, because it weighs trades instead of counting them. Win rate stays useful to understand a method's psychological comfort: losing six times out of ten demands a tolerance not everyone has, even when the system wins.
Because you are probably letting your winners run. A more distant target is reached less often, so win rate drops, but each win pays more, so RRR rises faster than win rate falls. That is the sign of a system improving, not the reverse.
In Tradoshi's Reports screen, win rate is computed as on this page: winning trades divided by winning and losing trades, with neutral trades, closed at zero, removed from the calculation. Their count is shown separately, in the same list.
On the Dashboard, the Win % gauge shows winners, losers and neutral trades at a glance, and the Day win % card gives the share of your positive days, which is the other reading described above. Sorting by tag, symbol or entry hour, in Reports, gives win rate per setup.
To go deeper on this: read the full guide. Every term is gathered in the glossary.