Guide · Risk / reward
The short answer
There is no good ratio in isolation. A ratio is only good relative to the win rate it needs to break even.
A risk-reward ratio is only meaningful against the win rate it needs. Any ratio can be excellent or worthless depending on how often the trade actually works, and the ratio alone tells you nothing about that.
What a ratio does fix is the break-even win rate — the frequency at which the strategy makes exactly nothing. Below it you lose; above it you gain. That number falls straight out of the ratio:
| Ratio | R | Break-even win rate |
|---|---|---|
| 1:0.5 | 0.5 | 66.7% |
| 1:1 | 1.0 | 50.0% |
| 1:1.5 | 1.5 | 40.0% |
| 1:2 | 2.0 | 33.3% |
| 1:3 | 3.0 | 25.0% |
| 1:5 | 5.0 | 16.7% |
Read this as a demand rather than a promise. A 1:3 trade needs to work only a quarter of the time — but you must actually reach the target a quarter of the time, and a target three times further away is reached less often than one nearby. Widening the target moves both numbers at once.
The planned ratio is measured to your target. The realised one is measured to your exits, and the two diverge for reasons that are entirely ordinary: you take a partial at 1.4R, you move the stop to break-even and get tagged out, the spread eats a fifth of a tight stop.
The honest version is to compute the ratio from closed trades, not from intentions. A trading journal that records both the planned and realised R is the cheapest way to see the gap.
Where this goes wrong
Costs shift every row of that table upward. On a 10-pip stop with a 1-pip spread, a nominal 1:2 is a real 1:1.8, and the break-even win rate moves from 33.3% to 35.7%. On tight stops the spread is not a rounding error, it is a material part of the trade.
The ratio should be an output of where the levels are, not an input you impose. Working backwards from a target you picked because it gives a tidy 1:3 is how a stop ends up somewhere the market has no reason to respect.
A ratio derived this way varies trade to trade, and that is correct — it reflects the structure actually in front of you. A ratio fixed in advance forces one of the two levels to be arbitrary, and it is almost always the stop.
The break-even table says where a ratio stops losing. This says what it earns above that line, in R per trade, which is the number that compounds.
| Win rate | 1:1 | 1:2 | 1:3 |
|---|---|---|---|
| 30% | −0.40R | −0.10R | +0.20R |
| 40% | −0.20R | +0.20R | +0.60R |
| 50% | 0.00R | +0.50R | +1.00R |
| 60% | +0.20R | +0.80R | +1.40R |
Read down a column rather than across a row: ten points of win rate is worth 0.2R at 1:1 and 0.4R at 1:3. The payoff multiplies whatever edge the win rate provides, which is why a small improvement in either is worth more at a higher ratio.
Taking half off at 1R and letting the rest run to 3R does not produce a 1:3 trade. It produces a blended one, and the blend is usually worse than either leg looks.
That is a legitimate choice — it lowers variance and makes a strategy easier to hold — but it has to be priced in. Compute the break-even win rate on the blended ratio, not the target, or the strategy quietly needs a higher hit rate than you think.
A 1:3 derived from a stop at the swing low and a target at the prior high is a description of the chart. A 1:3 produced by moving the stop closer until the arithmetic worked is a description of the trader.
Both look identical in a journal, which is why recording the reason for each level — not just its price — is what makes the ratio interpretable later.
None in isolation. What a ratio fixes is the win rate needed to break even: 1 ÷ (1 + R). A 1:1 needs 50%, a 1:2 needs 33.3%, a 1:3 needs 25% — all before costs, which raise every figure.
Not by itself. A target twice as far away is reached less often, so widening it lowers both the required win rate and the achieved one. The two move together.