What Is an R-Multiple?
An R-multiple is a trade's profit or loss divided by the amount that was at risk when it was opened. A trade risking $400 that made $1,200 is +3R. Because every trade is measured in the same unit, results across different account sizes and instruments become directly comparable.
How is it calculated?
R = (exit price − entry price) ÷ (entry price − stop price), with the sign flipped for shorts. The denominator — the initial risk — is fixed at entry and never changes, which is the entire point.
Worked example. Buy at $100.00, stop at $96.00, so the initial risk is $4.00 per share. Exit at $112.00 and the gain is $12.00 per share, which is +3.0R. Stopped out instead and the result is −1.0R. Exit early at $102.00 and it is +0.5R.
Position size does not appear anywhere in that arithmetic, which is why R-multiples let you compare a 100-share stock trade against a two-lot futures trade against an options spread.
Why measure in R rather than in dollars?
Because dollars conflate two decisions: whether the trade was good and how big it was. A $5,000 winner on a $50,000 risk is a worse trade than a $500 winner on a $100 risk, and only the R-multiple says so — +0.1R against +5.0R.
It also makes expectancy portable. A system averaging +0.35R per trade earns 0.35 times whatever you risk. Over 200 trades at 1% of the account per trade, that is 70R, or roughly 70% of starting equity before compounding. Change the risk per trade and the R figures do not move.
What does R not capture?
Slippage on the stop. A stop at $96.00 filled at $95.20 is −1.2R, not −1.0R. Recording planned R rather than realised R quietly flatters every losing trade.
Trades with no stop. If there was no initial risk defined, there is no denominator, and any R-multiple attached afterwards is invented.
Moving the stop. R is measured against the initial risk. Widening a stop mid-trade and then computing R against the new one turns a −2R loss into a −1R loss on paper. This is the most common way an R-based record is inflated.
What is a good average R?
Anything reliably above zero, over enough trades to be distinguishable from luck. A system with a 40% win rate averaging +2.5R on winners has an expectancy of 0.4 × 2.5 − 0.6 × 1.0 = +0.40R. One with a 60% win rate averaging +0.8R has 0.6 × 0.8 − 0.4 × 1.0 = +0.08R, which is positive and will be swamped by costs.
The sample matters more than the average. Thirty trades tells you very little; two hundred starts to.
How does this connect to a track record?
Expressing results in units of risk rather than in currency is the same move Kelly made in 1956 when he framed betting size as a fraction of bankroll: normalise by what was at stake and outcomes from different-sized positions become comparable.[1] An R-multiple is that normalisation applied per trade.
Normalising by risk also exposes the behaviour that currency hides. Odean found investors realise gains far more readily than losses[2] — a pattern invisible in a dollar P&L and obvious in a distribution of R-multiples, where the winners cluster small and the losers have a tail.
A metric is a summary of a record, so it inherits every weakness of that record. Computed from trades selected after the fact, it is a number about the selection. kappi commits each trade before it resolves and publishes it on a Merkle-anchored log, so PnL, RME, correlation to SPX, mean R:R and trade count over 30, 100 and 200-day windows are computed over everything that was committed, losses included. $15/month to keep a record; reading one is free.
Sources
Frequently asked questions
What does 3R mean in trading?
A profit of three times the amount risked on that trade. If the initial risk was $400, a +3R trade made $1,200. A trade stopped out at the original stop is −1R by definition.
How do you calculate an R-multiple?
Divide the trade's result by the initial risk: (exit − entry) ÷ (entry − stop). Buying at $100 with a stop at $96 and exiting at $112 is $12 ÷ $4 = +3R.
What is a good average R-multiple?
Any average reliably above zero over a large enough sample. A 40% win rate averaging +2.5R gives an expectancy of +0.40R per trade; a 60% win rate averaging +0.8R gives only +0.08R.
Can R-multiples be gamed?
Yes — by widening the stop mid-trade and computing R against the new one, or by recording the planned stop rather than the slipped fill. R must always be measured against the risk defined at entry.