What has been said publicly
In his three-setups article, Kullamägi writes that a trader can be very profitable with a win rate of only 25 to 30%, provided losses stay small and winners are large, and he frames the goal as making several times the initial risk on the trades that work. His public FAQ lists a 25% win rate for 2019 and says most trades risk around 0.3 to 0.5% of the account, rarely more than 1%.
Measure every trade in R
R is the amount you planned to lose if the stop is hit. A trade risking $250 that makes $1,250 is a 5R winner; a stop-out at the planned level is -1R. Measuring in R makes trades comparable across position sizes and account values.
Planned risk (1R)
Entry minus stop, times shares, set before the trade.
Result in R
Profit or loss divided by the planned risk.
Distribution
The spread of R results across a large sample, not one trade.
Expectancy in one line
Expectancy = (win rate x average win) - (loss rate x average loss). With a 25% win rate and losses held to 1R, the average winner has to exceed 3R just to break even. At an average winner of 5R, the expectancy is 0.25 x 5 - 0.75 x 1 = +0.5R per trade.
- Average winner 3R at a 25% win rate: break even, before costs.
- Average winner 5R: about +0.5R per trade.
- Average loss creeping to 1.5R with 5R winners: about +0.13R per trade.
The last line is the important one. Letting the average loss slip from 1R to 1.5R removes most of the edge, even though the win rate has not changed. This is why the stop discipline described in the setup material carries so much weight.
What losing streaks look like
At a 25% win rate, long runs of losses are normal. In a simple simulation of 100 independent trades, the median longest losing streak was 12 trades, and roughly 1 in 10 sequences had a streak of 18 or more. About 79% of sequences included at least one run of 10 straight losses.
At 0.5% risk per trade, 12 straight losses cost about 5.8% of the account. At 2% per trade, the same streak costs more than 21%. The win rate is the same; the risk per attempt decides whether the streak can be survived.
Where the math breaks down
- Gaps and slippage push real losses beyond 1R.
- Cutting winners early lowers the average win below break-even.
- Raising size after a streak concentrates risk at the worst time.
- Small samples can show any win rate; judge a few hundred trades.
- Commissions and borrow costs reduce every R result.
The expectancy review
- Record planned R before every entry.
- Log the realized R, including slippage.
- Track average loss separately; it should stay near -1R.
- Track the average winner and the share of trades above 5R.
- Review the longest losing streak against your risk per trade.
- Recalculate expectancy by setup and by market regime.