Last spring I looked at a friend's journal and found a 62% win rate. Sixty-two. He was proud of it. His account was down 14% on the year.
He wasn't unlucky, and he wasn't bad at reading charts. He was just getting paid the wrong amount for being right.
The arithmetic nobody does
Take a simple case. You win 60% of your trades, and the average winner makes $50. The other 40% lose, and the average loser costs $100.
Per trade, that's 0.6 × $50 minus 0.4 × $100. Thirty dollars in, forty out. You lose $10 every time you click, and the more you trade, the faster it adds up. Traders call that number expectancy: the average amount you make or lose per trade over a long run. Win rate is only half of it.
Flip the profile and it gets stranger. A system that wins 30% of the time, with winners averaging $250 and losers $50, makes 0.3 × $250 minus 0.7 × $50. That's $75 minus $35, so $40 per trade. Wrong seven times out of ten, and still comfortably profitable.
Our own BTC flip strategy sits at 21.9% wins and returned +4,909% over six years on Bybit. Nobody would call that a high hit rate. The winners are just much bigger than the losers.
Why the brain gets this backwards
Winning feels good and losing hurts, and we remember the count, not the size. A day with five small wins and one big loss feels like a good day, right up until you check the balance.
There's a mechanical reason too. Taking profit early is the easiest habit in trading. The trade is green, you're nervous, you close it. Cutting losers is the opposite: you hold, you hope, you move the stop. Do both for a year and you've built a high win rate and a negative expectancy, almost by accident.
I did exactly this for about two years. My win rate looked great. My equity curve looked like a slow leak.
What most traders track instead
Most traders track win rate because it's the number the platform shows first. It's also the easiest one to improve by cheating yourself: tighten the profit target, widen the stop, and the percentage climbs while the account shrinks.
A better habit is to track the ratio of average win to average loss next to your win rate, always together. Neither means much alone. A 40% win rate needs winners at least 1.5 times the size of losers just to break even, before fees. A 70% win rate can survive losers that are 2.3 times the size of winners, but not much beyond that.
That's the break-even line: loss size divided by win size plus loss size. Everything above it is edge. Everything below it is a hobby that costs money.
Here's the part that stings. Fees and slippage hit every trade, winners and losers alike. A strategy that looks flat on paper is usually negative live, so you want a cushion above break-even, not a hair over it. I like to see expectancy at least a few dollars per trade positive before I trust anything.
Do this on Sunday
Export your last 100 trades. Write down three numbers: win rate, average win, average loss. Then run the multiplication from earlier. It takes five minutes, and it's the least flattering and most useful five minutes of your trading week.
If the result is negative, don't tweak entries. Look at how you exit. In my experience the problem lives there far more often than in the signal.
If you'd like to see what a low win rate and a big payoff ratio look like across thousands of trades, the full data and methodology are at v33systematic.com.