![]()
Win rate is the percentage of closed trades that end with a net profit, calculated as winning trades divided by all closed trades. It is also called the winning rate or percent profitable. The figure shows how often a trader is right, not how much is earned when right, so it cannot show by itself whether a strategy makes money.
The most common misreading is that a higher win rate means a better strategy. A strategy that wins 80% of its trades loses money when its average loss is large enough, and one that wins 35% can be profitable when its average win is large enough. The reliability of a win rate also depends on sample size: 12 wins in 20 trades gives a 95% confidence interval of roughly 39% to 78%.
What Is Win Rate?
The formula is win rate = winning trades ÷ closed trades × 100. Only closed trades enter the calculation. Treatment of break-even trades varies, so the counting rule should be stated.
Consider a hypothetical sample of 20 closed trades in US stocks and BTCUSDT: 12 net winners, 7 net losers, and 1 break-even. Before fees, 14 of the 20 showed a gain, because two trades earned no more than the fees paid on them. The win rate depends on the counting rule.
All three figures describe the same trades. When break-even trades stay in the denominator, each one lowers the win rate slightly, which is why the second figure is higher. The third is higher because fees are ignored.
What Counts as a Winning Trade?
In this post, a winning trade is a closed trade whose realized profit is positive after all fees. The SEC's Office of Investor Education and Advocacy advises checking which fees and expenses a performance calculation includes, because fees and expenses reduce returns (SEC, Investor Bulletin: Performance Claims). Net P&L is covered in more detail in what P&L means in trading.
A break-even trade has a net result of zero, or one inside a tolerance fixed before the analysis. A trade is one position from flat to flat. When a position is built in several entries and closed in several exits, the whole sequence counts once, with its combined net result. Counting each partial exit as its own trade changes both the trade count and the win rate.
How to Calculate Win Rate
Four steps produce a win rate that can be compared across periods.
- Define each trade as one position from flat to flat, however many entries and exits it contains.
- Record each trade's net P&L. For a long, that is exit value minus entry value minus all fees. For a short, it is entry proceeds minus exit cost minus all fees.
- Classify each closed trade as a win if the net P&L is positive, a loss if it is negative, and break-even otherwise.
- Divide the wins by the closed trades, and record the trade count and the period beside the result.
A hypothetical US stock trade shows the arithmetic. A trader buys 100 shares at 50.00 USD and sells them at 52.50 USD, paying 2.00 USD in total fees. The net result is 250.00 − 2.00 = +248.00 USD, so the trade is a win.
A hypothetical crypto trade shows why fees matter. A trader buys 0.02 BTC on BTCUSDT at 60,000 USDT and sells at 60,100 USDT, with an assumed fee of 0.10% of trade value on each side. The gross gain is 2.00 USDT and the fees total about 2.40 USDT. The net result is about −0.40 USDT, so the trade is a loss even though the price rose.
Why a High Win Rate Can Still Lose Money
Win rate counts trades but ignores their size. The missing input is the payoff ratio, the average winning trade divided by the average losing trade. Win rate and payoff ratio together determine expectancy.
payoff ratio = average win ÷ average lossexpectancy = (win rate × average win) − (loss rate × average loss)
Expectancy is the average net result per trade, the expected value of one trade's outcome.
Setting expectancy to zero gives the break-even win rate: break-even win rate = 1 ÷ (1 + payoff ratio), assuming no break-even trades, so the loss rate equals one minus the win rate. The table applies it before fees.
The table is arithmetic, not a forecast. It shows the lowest win rate at which a strategy breaks even if its averages hold. The planned version of the payoff ratio is the risk-reward ratio, covered in the risk-reward ratio guide and the risk/reward ratio calculator.
Fees raise the bar. A hypothetical cost of 5 USD per trade turns a strategy that averages 100 USD per win and 100 USD per loss into one that averages 95 USD per win and 105 USD per loss. The break-even win rate then rises from 50.0% to 52.5%.
Two hypothetical traders show the effect. Each closes 100 trades, and fees are ignored.
Trader A is right four times in five and still loses money, because the average loss is five times the average win. Trader B is wrong in 65 of 100 trades and finishes ahead. Profit factor, gross profit divided by gross loss, combines the same inputs into one ratio and points the same way: below 1 for Trader A and above 1 for Trader B.
How Many Trades Does a Win Rate Need?
A win rate is an estimate, and its error is large when few trades stand behind it. The table shows the 95% Wilson score interval around an observed win rate of 60% at five trade counts. These figures are calculations from the formula, not market data.
The Wilson interval is (p + z²/(2n) ± z × sqrt(p × (1 − p)/n + z²/(4n²))) / (1 + z²/n), where p is the observed win rate, n is the number of trades, and z is 1.96 for 95% confidence.
The NIST/SEMATECH e-Handbook of Statistical Methods describes this method for a proportion, alongside an exact binomial method. It notes that symmetric limits based on the normal approximation may not be accurate enough when the sample is very small (NIST).
Brown, Cai and DasGupta recommend the Wilson or Jeffreys interval for samples of 40 or fewer, and report that the simple Wald interval has erratic coverage (Brown, Cai and DasGupta, 2001). For 12 wins in 20 trades, the exact binomial method gives an interval somewhat wider than the Wilson result (36.1% to 80.9% against 38.7% to 78.1%).
Three readings follow from the table. First, 20 trades say little: a trader whose true win rate is exactly 50% would record 12 or more wins in 20 trades about one time in four, by a binomial calculation. Second, the interval narrows with the square root of the trade count, so quadrupling the sample roughly halves the width, from about 26 points at 50 trades to about 13 points at 200. Third, even at 100 trades, the lower bound sits only just above 50%.

The interval assumes independent trades with one fixed chance of winning. Mixed setups, shifting market conditions, and clusters of related trades break that assumption, so the real uncertainty can be larger.
A trading journal that stores net P&L for every closed trade can show the trade count beside each statistic. Trading Journl, for example, calculates win rate, payoff ratio, and expectancy from saved trades, and its trading skill analysis labels results by sample size.
Reliable vs. Risky Readings of Win Rate
A win rate is most informative when its counting rule, sample, and context are all visible.

The SEC's bulletin on performance claims warns about "cherry-picked" results that show only good periods or only profitable investments. It advises questioning any presentation that does not cover different market conditions (SEC, Investor Bulletin: Performance Claims). A win rate quoted without a count, a period, or an average loss gives a reader no way to check for either problem.
Limitations and Risks
Win rate counts closed trades only, so a losing position that stays open does not enter it. Odean (1998) analyzed trading records from 1987 through 1993 for 10,000 accounts at a large discount brokerage. The study is limited to stocks in the CRSP price file, which 6,380 of those accounts traded, and it measures gains and losses against the average purchase price with commissions included. Investors realized 14.8% of their paper gains but only 9.8% of their paper losses. On days when an account holding two or more stocks made a sale, 13,883 sales were at a gain and 11,930 at a loss. The positions kept showed a paper gain 79,658 times and a paper loss 110,348 times. A win rate built from the sales alone would not show the unsold losses.
Win rate also says nothing about the order of results. With a true win rate of 60% and independent trades, a run of five or more consecutive losses has about a 46% chance of appearing at least once in 100 trades. The figure comes from a run-length recursion that tracks the length of the current losing streak trade by trade, under the same independence assumption. A streak of that length is therefore not, by itself, evidence that a strategy has stopped working.
A win rate describes past trades. The SEC investor bulletin on performance claims states that past performance cannot predict how an investment strategy will perform in the future (SEC, Investor Bulletin: Performance Claims). A win rate measured in one market regime, such as a steady uptrend, may not carry over to another.
A Checklist Before Acting on a Win Rate
- Is each trade counted from flat to flat, with scale-ins and scale-outs combined?
- Is each trade classified by net P&L after all fees, with the break-even rule stated?
- Are the trade count and the period shown beside the win rate?
- Are average win, average loss, payoff ratio, and expectancy shown beside it?
- Does the win rate exceed the break-even win rate that the payoff ratio and fees imply?
- Is the interval around the win rate narrow enough for the decision at hand?
- Have open positions with large unrealized losses been reviewed separately?
- Has the win rate been checked across different periods and market conditions?
Frequently Asked Questions
What is a good win rate for day trading?
There is no universal good win rate, because the rate needed to break even depends on the payoff ratio and on fees. A strategy that risks 1 to make 2 breaks even at 33.3% before fees, while one that risks 2 to make 1 needs 66.7%. A better target is positive expectancy after fees over a sample large enough to trust.
Published outcome data on day traders offer context. An SEC investor publication dated 2005 says that day traders typically suffer severe financial losses in their first months of trading and that many never graduate to profit-making status. A study of day traders in Taiwan from 1992 to 2006 reported that less than 1% of the day trader population could predictably and reliably earn positive abnormal returns net of fees (Barber, Lee, Liu and Odean, 2014).
Key Takeaways
Win rate shows how often trades finish in profit, and it is most informative when read with payoff ratio, fees, and sample size.
- Win rate is winning closed trades divided by all closed trades, with a winning trade defined net of fees and break-even trades handled by a stated rule.
- The break-even win rate is 1 ÷ (1 + payoff ratio), before fees: 50.0% at 1 to 1, 33.3% at 2 to 1, and 66.7% at 0.5 to 1.
- Expectancy, (win rate × average win) − (loss rate × average loss), decides whether a high or low win rate makes money.
- Twelve wins in 20 trades is a 60% win rate with a 95% interval of roughly 39% to 78%. One hundred trades narrow it to about 50% to 69%.
- Closed trades hide open losses, and even at a true 60% win rate, a run of five or more losses has about a 46% chance of appearing within 100 trades.
When a decision rests on a win rate, record the sample size, the counting rule, and the payoff ratio beside it, so the decision can be reviewed later against what actually happened.
References
- Brown, L. D., Cai, T. T., and DasGupta, A., "Interval Estimation for a Binomial Proportion," Statistical Science 16(2), 101–133, 2001. doi:10.1214/ss/1009213286
- NIST/SEMATECH, e-Handbook of Statistical Methods, section 7.2.4.1, "Confidence intervals," retrieved October 1, 2026. itl.nist.gov
- Odean, T., "Are Investors Reluctant to Realize Their Losses?" The Journal of Finance 53(5), 1775–1798, 1998. doi:10.1111/0022-1082.00072
- Barber, B. M., Lee, Y.-T., Liu, Y.-J., and Odean, T., "The Cross-Section of Speculator Skill: Evidence from Day Trading," Journal of Financial Markets 18, 1–24, 2014. doi:10.1016/j.finmar.2013.05.006
- SEC Office of Investor Education and Advocacy, "Investor Bulletin: Performance Claims," September 15, 2022. Investor.gov
- SEC, "Day Trading: Your Dollars at Risk," page dated April 19, 2005, retrieved October 1, 2026. sec.gov
For reference only: The information in this post is provided for general reference and does not constitute investment advice or a solicitation to trade. Your responsibility: Final investment decisions and their outcomes are entirely your own. No warranty: This information is not guaranteed to be accurate and cannot be used as a basis for legal liability for any investment result.