R-Multiple Distribution: Why Your Average Hides the Trades That Actually Made You Money

“My average R is +0.4” sounds like a clean, single-number summary of a strategy's edge. It is a summary — of a spread of outcomes that, for most strategies, is not symmetric and not evenly distributed around that number. A handful of large winners can produce a positive average sitting on top of a majority of losing trades, and a handful of large losers can produce a negative average sitting on top of a majority of winners. The average alone cannot tell you which one you're looking at.

What the average actually collapses

Average R-multiple is the sum of every trade's result, expressed as a multiple of its initial risk, divided by the number of trades. That single division is where the shape of the data disappears. A strategy with 40 trades at −1R, 55 trades at +0.3R, and 5 trades at +6R produces a positive average — and looks, from that one number, similar to a strategy where 70 of 100 trades land near +0.5R with no large outliers at all. Those are two completely different trading styles with different failure modes, and the average makes them indistinguishable.

Mean vs. median: the fastest skew check

The gap between the mean R-multiple and the median R-multiple is the cheapest signal that the distribution is skewed. If the mean is comfortably above the median, the average is being pulled up by a small number of outsized winners, and most individual trades are below what the headline number implies. If the two are close, the results are more evenly spread and the average is a fair description of a typical trade.

Skew signal = mean R − median R

Neither number is wrong on its own — a strategy that's built around occasional large wins and frequent small losses is a legitimate style, sometimes called a trend-following or breakout profile. The problem isn't having that shape. It's not knowing you have it, and grading day-to-day performance against an average that only a rare trade type actually produces.

Why the shape changes what a losing streak means

A strategy with a right-skewed distribution — most trades small losses or small wins, a few large winners carrying the average — will produce long stretches where every closed trade looks unprofitable, simply because the trades that make the strategy work are rare by design. Judging that strategy by its last 15 trades, without knowing the distribution it was built on, looks exactly like judging a broken strategy. The opposite shape — most trades clustered near a small positive R with a few sharp losers — will look great for long stretches and then give back weeks of gains in one or two trades that were always part of the distribution, not a sign something suddenly broke.

Neither read is available from the average. Both are visible immediately once the individual R-multiples are plotted instead of collapsed.

What to actually look at

• A histogram of R-multiples, not a running average — bucket every closed trade by its result and look at the shape, not a single point estimate of it.
• The contribution of the top decile — what share of total R came from the best 10% of trades. A strategy where the top 10% of trades account for most of the positive R is not the same strategy as one where returns are spread evenly, even if the averages match.
• The worst single trade as a share of total R— the mirror question for the loss side: how much of the account's progress one bad trade can erase.
• Median R alongside mean R, every time the average is quoted — the two together take a few seconds to compare and immediately flag skew the average hides.

Reading the distribution, not the number

A distribution bunched tightly around a small positive median with a short right tail is a strategy that wins often and rarely by much — expectancy comes from frequency, and a string of losses is more likely to mean something changed. A distribution with a long right tail and a median near zero or slightly negative is a strategy that wins rarely and big — expectancy comes from the tail, and a string of small losses is the strategy working exactly as designed, not a signal to abandon it. Neither shape is right or wrong in the abstract. Trading one as if it were the other — cutting winners early because the average says +0.4R is “enough,” when the whole edge lives in the rare trade that runs to +6R — is where the distribution actually costs money.

Why this is hard to keep straight from memory

Recalling which recent trades were the small grinding wins and which were the rare large ones is not something most traders can do accurately after a few dozen trades, and a broker statement totals P&L, not R-multiples bucketed by size. Seeing the actual shape of a strategy's results means computing R for every trade against its own initial risk and plotting the full set, not the running average. getALPHAcomputes R-multiple for every logged trade automatically and keeps the full distribution alongside the summary stats, so the question isn't “what's my average R” but “what does my average R actually come from” — and that answer is what tells you whether a losing stretch is noise or a warning.