Expectancy answers the only question that matters about a strategy: if I take this trade a hundred times, do I end up ahead? A system with 45% wins at 1:2 has an expectancy of +0.35R, which is a real edge. The same win rate at 1:1 is -0.10R, which is a slow leak.
Expectancy is also fragile to costs. Fees, spread, funding and slippage all come out of the same number, and a thin edge can be eaten entirely by the venue.
The mistake it causes: abandoning a positive-expectancy system during a normal losing streak, which is the one outcome expectancy guarantees will happen regularly.
The formula, and what it answers
Expectancy = (win rate × R) − (1 − win rate), measured in R per trade. It answers the only question that matters about a strategy: if I take this trade a hundred more times, am I ahead?
| Win rate | R | Expectancy | Verdict |
|---|
| 70% | 0.3 | −0.09R | Feels great, loses money |
| 50% | 1.0 | 0.00R | Break even before fees |
| 45% | 2.0 | +0.35R | A real edge |
| 35% | 3.0 | +0.40R | A real edge, uncomfortable to trade |
The first row is the important one. A 70% win rate is the thing most traders chase, and paired with small targets it is a losing system that feels like a winning one for months.
What positive expectancy pays
At 1% risk on a $10,000 account, 1R is $100. An expectancy of +0.2R is $20 a trade, $2,000 over a hundred trades, before fees. That is roughly a year of patient swing trading for a 20% return, and it is what a genuinely good retail edge looks like. Anyone promising substantially more per month is selling something.
Expectancy is fragile to costs
Fees, spread, funding and slippage come out of the same number. A system at +0.1R before costs can be negative after them, which is why cost comparison is not a side issue for an active trader. It is the difference between an edge and a hobby.
You need a sample before you believe it
Around thirty trades before the shape is suggestive, closer to a hundred before it is evidence. Below that, luck dominates completely. And almost everyone overstates their own win rate from memory, which is why the number should come from a written record rather than a recollection.
