In crypto trading, a high win rate and a shrinking balance often sit side by side. The number of times you were right and the amount you gained are two different things.
What the profit factor is
It is the total amount gained on winning trades divided by the total amount given back on losing trades.
Profit factor = total of wins / total of losses
Say a period ended with 30,000 yen gained across the winning trades and 20,000 yen given back across the losing ones. The sum is:
1.5 = 30,000 / 20,000
Only amounts go into it; the number of wins does not.
At exactly 1.0 the amount gained and the amount given back are the same. The two cancel out and the balance ends where it started.
Above 1.0
The gains are the larger side. The 1.5 above means 1.5 was gained for every 1 given back.
In concrete terms, this is what that describes.
The losing trades took 20,000 yen in total. The winning trades brought in 30,000 yen in total. What is left in the account is the 10,000 yen difference.
Against 20,000 yen given back, 30,000 yen came in. For every 10,000 yen lost, 15,000 yen was recovered, and the 5,000 yen difference is what the balance grew by.
The important part is that this is not a statement about any single trade.
One position might return one and a half times its risk; the next might be lost in full. The ratio of 1.5 only appears once every closed position is added together.
What happens as the number of trades grows? If the same 1.5 holds, half of everything given back stays in the account.
By the time the losing side totals 200,000 yen, the winning side is 300,000 yen and 100,000 yen remains. At two million and three million the figure is one million. The relationship holds as long as the ratio does.
So the thing to watch is whether 1.5 is being held, not whether the last trade was a win. Losses do not have to be eliminated for the balance to grow.
The balance grows, and grows further the more positions are closed. That is the case for carrying on as you are.
Below 1.0
The losses are the larger side. A figure of 0.8 means only 0.8 was gained for every 1 given back.
Put in the same concrete terms:
The losing trades took 20,000 yen in total and the winning trades brought back 16,000 yen. The account is 4,000 yen lower than where it started.
By the time the losing side totals 200,000 yen, only 160,000 yen has come back, leaving a shortfall of 40,000 yen. The shortfall grows in proportion to the number of trades.
The balance falls, and repetition widens the gap rather than closing it. Nothing stops that except changing how the positions are opened and closed.
Even a figure of exactly 1.0 is negative in practice if it was worked out before fees.
What it shows, and what it is good for
What it shows is how much you gain for each unit you give back. It turns the efficiency of the amounts, rather than the count of correct calls, into a single figure.
It is useful in three places.
- Deciding whether to continue: above 1.0 keep going, below 1.0 change something
- Comparing: last month against this month, or one coin against another, on the same scale
- Finding what to fix: below 1.0, look at the size of the losses before the win rate
What it cannot tell you is when to trade or which coin to hold. It measures trades that are already finished and says nothing about the price from here.
Why this number is needed
Because the balance on its own does not tell you whether the way you are trading is worth continuing.
A balance moves with the market as well as with your decisions. A month when the coin you held happened to rise looks the same as a month when your approach was working.
This article explains what that number means and how to work it out.
It also shows how to work the number out from your own closed positions, using nothing more than the counts and the averages that most summary screens already show.
Every figure below is a worked example that anyone can reproduce with arithmetic. None of them is a result that someone achieved.
It is written for people trading crypto with their own funds who want to judge for themselves whether the way they are trading is worth continuing. It describes the situation as of 19 September 2026.
The win rate does not decide whether your balance grows
When you look back at a record, the win rate is the first thing you want to see. How many of the last ten closes were profitable. It is easy to read and it matches how the month felt.
The trouble is that the win rate does not decide whether the balance grows. If the amount per trade differs between wins and losses, the side with more occurrences can still be the losing side.
An example: 80 per cent right, and still down
Suppose you win eight times, gaining 1,000 yen each time, and lose twice, giving back 5,000 yen each time. The win rate is 80 per cent and the balance is 2,000 yen lower than when you started.
The balance fell because one loss was five times the size of one win. The count went one way and the size went the other, and size is what the account is made of.
The reverse also holds. A win rate of 30 per cent can grow the balance: three wins of 10,000 yen against seven losses held down to 1,000 yen each.
That comes to 30,000 yen gained against 7,000 yen given back, so the balance is 23,000 yen higher even though seven closes out of ten were losses.
Reading a record by the ratio of amounts
Put the two examples above through the formula from the opening of this article.
The 80 per cent example works out at 8,000 divided by 10,000, or 0.8. The 30 per cent example works out at 30,000 divided by 7,000, or 4.28.
The higher win rate scores 0.8 and the lower one scores 4.28. Reading by amounts rather than counts puts the two records in the opposite order.
The win rate and the profit factor move independently of each other. Neither one alone tells you whether to carry on, and it is the second of the two that decides whether the balance grows.
Use amounts after fees. On crypto perpetual futures there is also the funding rate, paid or received for as long as the position stays open.
A profit factor that leaves those out always looks better than the account does, and the more often you trade, the further the figure drifts from what the balance actually did.
How to work it out from your own record
Four numbers are enough: the number of wins, the average amount of a win, the number of losses and the average amount of a loss. Most screens that summarise closed positions show all four.
There is no need to add up the individual trades. Multiplying a count by an average gives the total for that side, which is all the division needs.
- Wins multiplied by the average win gives the total of wins
- Losses multiplied by the average loss gives the total of losses
- Divide the first by the second
The worked example in figure 2 comes to 0.02, far below the break-even line.
Repeating that sequence would keep taking the balance down: the fourteen losses cancel the three wins several times over.
Count closed positions only. Unrealised profit and loss on positions you still hold does not belong in the calculation.
An unrealised figure changes until you close, so mixing it in lets you pick a moment when the record looks better than it is.
Count one coin and one direction at a time. If several coins are mixed together the sizes are in different units, and the average amount stops meaning very much.
While the counts are small the number jumps around. One large loss can push it below 0.1 and one large win can lift it above 3.0.
It starts to read as a tendency somewhere past twenty closed positions.
What to look at when the number is below 1.0
When the profit factor is below 1.0, trying to raise the win rate is the long way round. The first thing to compare is the average loss against the average win.
If the gap is wide, the cause is either where you stop a losing position or the size you opened it with.
The average size on each side tells you which of the two it is. When the size is larger on the losing side, the positions opened to make something back are usually the ones doing it.
Evening out the size is a change you can make today. It does not depend on reading the market correctly, only on what you do when you open a position.
Holding time reads the same way. If losing positions are held for longer on average than winning ones, they are not being closed where you meant to close them, and you are waiting for a recovery.
When the figure is above 1.0 the first step is the same: look at the counts and check that the result is not one lucky position carrying the total.
If the counts are large enough and the figure still sits above 1.0, the approach is worth keeping. Whether to trade larger is a question for after that check, not before it.
✅ What tends to work
Narrow the record to one coin and one direction, then read the count, the average amount, the average size and the average holding time side by side. If the size differs between wins and losses, evening out the size is something you can change immediately.
⚠️ Common pitfalls
Trusting the number while there are only about five closed positions, and rebuilding the whole approach around it. At that count a single result flips the figure from one side of 1.0 to the other, and the change you make is a reaction to noise.
💡 To work out the profit factor on your own trades
The positions screen in Hype Trade splits your closed positions into wins and losses and shows the count and the average amount for each. Those four numbers are all the division above needs (as of 19 September 2026).
This article explains how to read a record of crypto trades. It is not advice to buy or sell, and the figures are worked examples rather than results anyone achieved. Fees and funding rates differ by venue and by coin.
Categories