How to Calculate Your Win Rate
To calculate your win rate, divide the number of winning trades by the total number of trades over a defined period. The figure is close to meaningless without the average win and average loss beside it, because those decide whether the record is profitable.
Win rate is winning trades divided by total trades. The calculation takes seconds. The difficulty is that the number is quoted constantly as though it measured skill, and on its own it measures almost nothing.
Before you start
A complete record of every trade, including the ones you would rather forget. A record missing the impulsive trades produces a figure describing a trader who does not exist.
A definition of what counts as a win, decided before counting. Is break-even a win. Is a trade closed for a tiny gain. Decide once, apply throughout.
An understanding that the figure means nothing without the average sizes alongside it. Those two numbers are what turn it into information.
The steps
1. Define a win before you count anything
Any positive result after costs is the usual definition. Break-even trades are typically excluded from both counts rather than assigned to one.
2. Count every trade in the period
Including the ones closed within a minute and the ones taken outside the plan. Those are disproportionately losses and disproportionately absent from most records.
3. Divide and stop there
Wins over total. The result is a percentage, and at this point it tells you nothing about whether you made money.
4. Compute the average win and average loss
Total gains divided by winning trades; total losses divided by losing trades. Those two figures with the win rate are what determine whether the record is profitable.
5. Combine all three into expectancy
Win rate times average win, minus loss rate times average loss. That is the expected result per trade, and it is the figure worth watching.
6. Break it down by setup
A single figure across three different setups is an average of three different things. Split by setup and the useful information appears.
7. Require enough trades before believing it
Twenty trades produces a figure that will move substantially with the next five. Treat anything under fifty as an early indication rather than a measurement.
How to tell it worked
A win was defined before counting, and applied to 100 percent of trades.
Every trade in the period is in the record, including impulsive ones.
Average win and average loss were computed alongside, not the rate alone.
And the sample covers at least 50 trades before any conclusion was drawn.
Why the figure can move the wrong way
Taking profits earlier raises it and lowers your results. More trades finish positive and each finishes smaller, so the rate improves while the total falls. The two measurements point in opposite directions.
Costs decide the borderline cases. On this site’s shared series a round trip measures about 2% of the median bar range of 0.493, which is enough to move a marginal trade from one column to the other.
What a high rate can conceal
Nine wins of one unit and one loss of twelve is a losing record at a win rate of 90%. The arithmetic is trivial and the presentation is not: the count says the method works and the currency says it does not.
Structures that produce high rates by design are common. Selling premium, fading small moves, taking quick profits — all produce long runs of small wins and infrequent large losses.
Which is why expectancy exists. It is the one figure that cannot be improved by trading in a way that loses money, and it is quoted far less often than the rate that can.
What the rate is genuinely useful for
Checking whether a method is behaving as designed. A trend approach that wins a minority of the time is working correctly; the same approach suddenly winning most of the time means something has changed and it is worth finding out what.
Detecting drift in your own execution. A rate that moves substantially without any change to the rules usually means the rules stopped being followed — cutting winners earlier, holding losers longer, or taking trades outside the plan.
Setting expectations before a losing run. A method winning four times in ten will produce runs of five losses regularly, and knowing the rate in advance is what makes that survivable rather than alarming.
None of those uses involve comparing it to anybody else’s. The figure is only meaningful against your own history and your own average sizes, and a rate quoted without a method attached to it describes nothing at all.
The original data
Of the 24,971 unique videos in research/search-study-corpus.jsonl, 161 mention win rates in the
title, at a median of 13,711 views across 130 channels, and 71% of those titles are
instruction-shaped. Risk management appears in 410 at 4,079 and drawdown in 7 at 3,813. The counts
come from site/corpus_count.py.
161 videos on win rate against 7 on drawdown. Twenty-three times the coverage for the statistic that can be improved by trading worse, and almost none for the one that measures what a losing stretch actually costs.
The answer to the question on that chart is that neither is wrong. A high proportion of small wins and a few large losses produces exactly that — and the account balance is the measurement that matters, with the rate explaining how it happened rather than contradicting it.
When it fails
The failure is optimising for the statistic, and it degrades results while every number improves. Profits get taken earlier, so more trades finish green. Losing trades get held slightly longer in the hope of recovery, so fewer finish red. Both changes raise the win rate. Both also shrink the average win and grow the average loss, which is the pair of numbers nobody was watching — and the account falls while the headline figure climbs.
The second failure is an incomplete record. The missing trades are mostly losses.
A third is no definition of a win. Borderline trades get assigned by mood.
A fourth is one figure across several setups. It describes none of them.
A fifth is a small sample. Twenty trades will move a lot.
And a sixth is quoting it without the sizes. Alone it cannot say whether you made money.
Related
Win rate covers the statistic and its limits. Trade review is where the record comes from. And trading journal is what makes the record complete enough to use.
I stopped quoting it on its own once I worked out that it can move in the opposite direction to my results. Taking profits earlier raises it and lowers what I make. It is a real measurement of a real thing and it is not the thing most people think it is measuring.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.