R-Multiple

R-multiples solve a fundamental problem in trade evaluation: comparing trades across different position sizes and account sizes. Without normalisation, a $500 win on a $50,000 account is the same as a $50 win on a $5,000 account, but raw dollar comparisons obscure this. R-multiples express both as their R equivalent.

Formal definition: R is the initial risk per trade in account currency, computed as |entry price − initial stop-loss| × position size × pip value. The R-multiple of a trade outcome is the realised P&L divided by R. A −1R trade lost exactly the planned risk; a +3R trade won 3× the planned risk.

Why R-multiples matter for strategy evaluation: - They make expectancy calculation account-size-independent - They allow direct comparison across different position sizes, instruments, and timeframes - They expose poor trade management: trades that should be +3R but realised as +0.5R indicate premature exits or trailing-stop logic that gives back too much - They isolate strategy edge from money management decisions

Practical use: tracking R-multiple distribution over many trades reveals the strategy's character. A trend-following system typically shows a long-tailed positive distribution (occasional +5R, +10R wins balancing many small losses). A mean-reversion system shows clustered +1R to +2R wins. A grid system shows clustered +0.5R wins with rare catastrophic −10R or worse losses.

Average R-multiple for evaluation: a strategy averaging +0.3R per trade with 100 trades/year produces 30R annual return. If risk per trade is 1% of account, that translates to 30% annual return (before compounding). This decomposition makes strategy evaluation modular: average R per trade × trades per year × % risk per trade = annual return.

Limitation: R-multiples assume the initial stop-loss is the relevant risk, which fails for strategies that close at adverse price changes before the stop is hit (e.g. partial closes, time-based exits). In those cases, realised R may understate the actual risk taken.