Why a 40% win rate can be very profitable
Lesson 12 · about 8 min
Beginners chase win rate because winning feels like being right. This lesson walks through a 40% system trade by trade, to show why its arithmetic works and why its emotions do not, so that you can hold onto it when the losses cluster.
The system
- Win rate: 40%.
- Average winner: 2.5R.
- Average loser: 1.0R.
Expectancy = (0.40 × 2.5) − (0.60 × 1.0) = 1.00 − 0.60 = +0.40R per trade.
Over 100 trades that is +40R. At 1% risk, roughly +40% on the account over those 100 trades, before costs. Over the same 100 trades the system loses 60 times.
Compare it with an 80% win-rate system whose winners average 0.3R and whose losers average 1.5R (because the trader holds losers hoping they recover):
Expectancy = (0.80 × 0.3) − (0.20 × 1.5) = 0.24 − 0.30 = −0.06R.
Eighty wins out of a hundred, and the account is smaller at the end.
Key idea: Win rate is one of four inputs to expectancy, and not the most important one. The ratio of average winner to average loser decides how low the win rate is allowed to be.
Ten trades, in order
The averages hide what it feels like to live inside the system. Here are ten trades in a realistic order: four wins at +2.5R, six losses at −1R.
| Trade | Result | Running total |
|---|---|---|
| 1 | −1.0R | −1.0R |
| 2 | −1.0R | −2.0R |
| 3 | −1.0R | −3.0R |
| 4 | +2.5R | −0.5R |
| 5 | −1.0R | −1.5R |
| 6 | −1.0R | −2.5R |
| 7 | +2.5R | 0.0R |
| 8 | −1.0R | −1.0R |
| 9 | +2.5R | +1.5R |
| 10 | +2.5R | +4.0R |
Ten trades, +4R, exactly the expectancy (10 × 0.4R). But look at the path. The account was underwater after eight of the ten trades. It hit −3R after three straight losses at the start. A trader who had not done the arithmetic in advance would have concluded by trade 3 that "the system does not work", changed something, and never collected trades 9 and 10.
This is the practical reason to compute expectancy and streak length before you start. Not because the numbers change anything about the market, but because they change what you do at trade 3.
What the streaks look like
With a 60% loss rate, the chance that any given run of trades is six straight losses is 0.6^6 = 4.7%. Over 100 trades, expect that to happen roughly four or five times. Seven straight: 0.6^7 = 2.8%, roughly two or three times per 100 trades. Nine straight: 1.0%, about once.
| Straight losses | Probability at any point | Expected occurrences per 100 trades | Drawdown at 1% risk |
|---|---|---|---|
| 4 | 13.0% | ~13 | 3.9% |
| 6 | 4.7% | ~4 to 5 | 5.9% |
| 8 | 1.7% | ~1 to 2 | 7.7% |
| 10 | 0.6% | ~0 to 1 | 9.6% |
At 1% risk, even the ten-loss streak is under a 10% drawdown, comfortably on the flat part of the recovery curve. At 3% risk the same streak is a 26% drawdown and the trader is at the bend, right when they most need to keep taking the next signal. Position size is what makes a low-win-rate system tradeable at all.
Where the 2.5R comes from
A 2.5R average winner does not mean every winner is 2.5R. It typically means most winners are 1.5R to 2R and a few are 4R to 6R. Those few big winners are the profit. Rules that protect them:
- Do not take profit at 1R "to lock something in". Doing so caps every winner and turns the system into a 40% win rate with 1R winners, which has expectancy 0.40 − 0.60 = −0.20R.
- Do not widen the stop on losers. Every widened stop turns a −1R into −1.5R or −2R and eats a big winner.
- Accept that many trades that reach +1R will come back and stop out at −1R. That is the cost of holding for the 4R.
The emotional contract
Write this down before you trade a system like this, and re-read it during the losing streaks:
"I expect to lose about 6 trades in 10. I expect to see 6 straight losses several times a year. As long as my losers average close to 1R and my winners average above 2R, the system is working exactly as designed. I will judge it on 100 trades, not on this week."
Try it: Simulate the system with a coin or a spreadsheet: random number under 0.40 is a +2.5R win, otherwise a −1R loss. Run 100 trades, plot the running total, and count the longest losing streak. Do it three times. Notice how different the paths look and how similar the endpoints are.
Recap
- 40% win rate with 2.5R winners and 1R losers has an expectancy of +0.40R; 80% win rate with 0.3R winners and 1.5R losers loses money.
- Inside a profitable low-win-rate system, being underwater for most of a ten-trade run is normal.
- Expect six-loss streaks several times per 100 trades; at 1% risk they cost about 6%.
- The profit is in the few large winners; taking profit early or widening stops destroys the edge.
- Decide in advance how many trades you will judge the system on, and hold to it through the streaks.
See it drawn
Original diagrams for the ideas on this page. Illustrative, not real market data.