Win rate is the percentage of your trades that close green, and on its own it tells you almost nothing about whether a system makes money. A 30% win rate can be highly profitable and a 70% win rate can be ruinous, because the number says nothing about the size of the wins or the losses.
It is the first statistic new traders ask about and the first one sellers of trading products advertise, for the same reason: it sounds like a measure of skill and it is easy to say out loud. It is neither. Win rate is one of two inputs into whether a strategy works, and it is the less important one.
What win rate actually measures
Win rate answers one question: out of the last hundred trades, how many closed above the entry price after costs? That is all. It does not know whether the winners were worth twenty dollars or two thousand, whether the losers were capped or open-ended, or whether the trades were even the same size as each other.
Because it ignores magnitude, win rate can move in the opposite direction to your account. Every trader who has taken profit at half a target while letting a loser run past its stop has raised their win rate and lowered their balance in the same afternoon.
The breakeven win rate for each reward-to-risk ratio
A win rate only becomes meaningful next to the average size of a win relative to a loss. For a system where every winner returns R times the amount risked, the win rate needed just to break even is 1 ÷ (1 + R). That single formula collapses most arguments about win rate.
| Average reward-to-risk | Breakeven win rate | What a 45% win rate means here |
|---|---|---|
| 0.5 : 1 | 66.7% | Losing badly |
| 1 : 1 | 50% | Slightly losing |
| 1.5 : 1 | 40% | Profitable |
| 2 : 1 | 33.3% | Comfortably profitable |
| 3 : 1 | 25% | Very strong |
| 4 : 1 | 20% | Exceptional |
Read the last column. The same 45% win rate is a disaster in the top row and an outstanding result in the bottom one. Nothing about the trader changed between those rows — only the shape of the exits. This is why the reward-to-risk ratio belongs in the same sentence as any win rate, always.
Your exits set your win rate, not your edge
Here is the uncomfortable part. Win rate is far more a record of your exit discipline than of your ability to pick direction, and it can be manufactured on demand. Widen your stop and take profit early, and the percentage goes up while the expectancy goes down. Do the reverse and the percentage falls while the account improves.
The tendency is well documented. Terrance Odean's study of ten thousand retail brokerage accounts found investors were roughly 50% more likely to sell a position showing a gain than one showing a loss, and that the winners they sold went on to outperform the losers they kept (Odean, "Are Investors Reluctant to Realize Their Losses?", Journal of Finance, 1998). That behaviour produces a flattering win rate and a shrinking account at the same time — the two numbers are not just different, they are being pushed apart by the same habit.
Which is why a trader who wants a better win rate can get one this week, and it will not help. The fix runs the other way: define the exit before the entry, then let the win rate be whatever the market makes it. That is the argument behind taking partials without capping your winners.
How a very high win rate hides fragility
Strategies advertised at 90% and above almost always share one structural feature: a small, frequent gain funded by a rare, very large loss. Selling premium without protection, averaging into a losing position, and trading without a hard stop all produce beautiful win rates for months and then a single trade that erases them.
- No stop means no maximum loss. A win rate is only interpretable if the loser is bounded. Without a defined invalidation, the average loss is not a number you know — it is a number the market will tell you later.
- Recovery is asymmetric. One loss of ten units after nine wins of one unit leaves you flat. The win rate reads 90%.
- Small samples flatter. Twenty trades is not a track record. A coin-flip system produces a 70% "win rate" over twenty trades often enough that it proves nothing.
- Streaks scale with win rate, in both directions. A 40% system will hand you a run of six or seven losers eventually. That is arithmetic, not a broken strategy — see surviving a losing streak.
Where win rate does genuinely matter
None of this makes the number useless. It matters in three specific places, all of them about survivability rather than profitability.
- Streak length. The lower the win rate, the longer the losing runs you must be sized and psychologically prepared for. This is a real constraint on how much you can risk per trade.
- Feedback speed. A higher win rate gives you more frequent confirmation that the process is working, which makes a plan easier to follow. That is a behavioural advantage, and it is worth something.
- Sample size needed. Systems with very low win rates and very large winners need far more trades before their results mean anything, because the outcome depends on catching the rare large move.
The disclosure most traders never look up
If you want to see how far win rate sits from profitability at industry scale, the data is public and required by law. Under 17 CFR 5.5, US retail foreign exchange dealers must disclose, for each of the four most recent calendar quarters, the total number of non-discretionary retail forex accounts they hold and the percentage of those accounts that were profitable and not profitable — with break-even accounts counted as not profitable. It is on the risk disclosure statement your broker hands you.
Those percentages describe accounts, not trades, which is exactly the point: they measure whether customers finished the quarter up, and the answer for most firms is that a minority did. Individual trades close green far more often than accounts do. For the wider picture, see what percentage of day traders are profitable.
Frequently Asked Questions
What is a good win rate in trading?
There is no good win rate in isolation, because the number only means something next to your average reward-to-risk. At 1:1 you need to win more than half your trades to break even. At 3:1 you need 25%. A 40% win rate is excellent on a 3:1 system and fatal on a 0.5:1 one, so quoting a win rate without the ratio beside it is meaningless.
Can you be profitable with a 30% win rate?
Yes. A system that loses seven trades out of ten and makes three times its risk on the three winners returns nine units of profit against seven units of loss before costs. The cost of that structure is psychological rather than mathematical: long losing runs are normal at a 30% win rate, so it only works if your position size and your temperament can absorb them.
Why does my win rate look better than my account?
Almost always because of how you exit. Closing winners quickly and holding losers in the hope they come back raises the percentage of trades that finish green while making the average loss bigger than the average win. Win rate is largely a record of your exit habits, not a measurement of how good your entries are.
What should I track instead of win rate?
Track expectancy, which combines win rate with the average size of wins and losses into the amount you expect to make or lose per trade. Alongside it, track average win divided by average loss, maximum drawdown, and the longest losing run in your record. Win rate is still worth logging, but only as one input into those numbers.
Bottom line
Win rate is a headline number that answers a question nobody should be asking on its own. Pair it with the average reward-to-risk and it becomes useful; quote it alone and it is closer to marketing than measurement. The number that actually settles whether a system works is expectancy, and the framework it sits inside is risk management in trading. Log your win rate, by all means — then stop treating it as a scoreboard.