R-Multiples: Measure Every Trade in Risk, Not Dollars

Two traders send you their month. The first made $412. The second made $1,180. You know almost nothing about either of them.

You do not know what they risked to get there, how many trades it took, or whether the good month came from the strategy or from one oversized position that happened to work. Dollars answer how much, and nothing else. R-multiples answer the question the dollar figure hides, which is whether the decisions were any good.

This is what an R-multiple is, how to compute one correctly, the choice of denominator that quietly decides what your number even means, and the four ways an R column turns into fiction without anyone noticing.

The Definition, and the Part People Skip

R is the risk on the trade. Entry price to stop price, multiplied by position size. Fix that number before you enter and it becomes the unit you measure the result in.

An R-multiple is the result divided by R.

A trade entered at 1.0850 with a stop at 1.0820 risks 30 pips. Exit at 1.0910 and the profit is 60 pips, so the trade is 60 divided by 30, or +2.0R. Exit at 1.0835 for 15 pips and it is +0.5R. Get stopped at 1.0820 and it is -1.0R, which is true by construction: the loss is exactly the thing in the denominator.

The label comes from Van K. Tharp, who popularised it in Trade Your Way to Financial Freedom in the 1990s. Normalising results by risk is older than the name, but the name is what made it standard among retail traders.

The part people skip is that both halves have to be in the same unit. Pips over pips works. Dollars over dollars works. Dollars of profit over pips of stop distance is not an R-multiple, it is a number.

What the Unit Buys You

Here are two blocks of ten trades from the same trader running the same setup, one on a small personal account and one on a funded account.

In dollars the funded account won, by half again. In R the personal account produced four times the edge, and the funded account barely broke out of noise.

The dollar column is not lying. It is answering a different question, which is how big the account was. That is worth knowing at the end of the month and useless when you are deciding which of the two ways you were trading is worth repeating.

R is what lets you add unlike trades together. A gold trade and a EUR/USD trade have nothing in common in pips or in dollars. In R they are directly comparable, which is what makes an average, an expectancy figure or a profit factor mean anything at all across a mixed log.

The Denominator Decides the Question

This is the part almost nothing written about R-multiples makes explicit, and it is where two honest traders end up with two different numbers for the same trade.

There are two defensible denominators.

Per-trade R. Divide each trade by the risk that trade actually carried. Every full stop-out is -1R, every trade is measured against its own bet. Position sizing disappears completely.

Fixed-unit R. Decide that one R is a constant, say 1 percent of the account, and divide every trade by that. A trade you sized at 2 percent that lost is -2R. Sizing stays visible.

Per-trade R evaluates the strategy. Fixed-unit R evaluates the trader. Most journals, including this one, use per-trade R, because the strategy question is the one people ask first. That default has a consequence worth seeing on paper.

Take a hundred trades, all the same setup, split by how confident the trader felt:

The R column says the setup made 18R over a hundred trades. The account is down 3 percent. Nothing here is a rounding error, and no trade was logged wrongly. Conviction was uncorrelated with outcome, the large bets went to the losing half, and per-trade R deleted exactly the information that explains the loss. (Sizing is off a flat starting balance so the arithmetic stays readable; compounding moves these figures slightly, not the conclusion.)

The lesson is not that R is broken. It is that R and the equity curve answer different questions and you need both open at once. When they disagree, the disagreement is the finding: your setup works and your sizing is eating it, or the reverse.

Four Ways an R Column Turns Into Fiction

1. Planned R and realised R in the same column

Planned R is entry to target divided by entry to stop, and it is known before the trade. Realised R is entry to exit divided by entry to stop, and it is known after. They are different numbers, and the planned one is almost always the larger.

Plan a 3R trade, take profit early at 1.4R, log the plan, and your average win reads 3R. Do that on a third of your trades and every downstream number, expectancy included, describes a trader who does not exist.

Pick one and keep the column pure. Realised R is the one that belongs in the performance stats. Planned R is worth logging too, in its own field, because the gap between the two is a real and measurable habit: cutting winners early shows up as a persistent shortfall between planned and realised on trades that never touched the stop.

2. The stop you moved

You enter at 1.0850 with the stop at 1.0820, 30 pips of risk. It goes against you, you widen the stop to 1.0790, and it comes back for a 60 pip win.

Against the planned stop that is +2.0R. Against the stop that was actually live it is +1.0R. Both are arithmetically correct, and the first is the one your journal will record if it divides by the original number.

The loss case is worse. Had it stopped at the widened level, you would have lost 60 pips, which is two units of the risk you had budgeted. A per-trade R column records it as -1.0R, identical to a disciplined loss. The single most expensive habit in retail trading leaves no trace in the statistic that exists to measure risk.

The fix is not clever maths. Log the trade against the widest stop it ever carried, so the win reads +1.0R and the loss reads -2.0R in budget terms, and flag it as a rule break so it lands in your discipline score rather than quietly inflating your edge. If you find yourself doing it often, the cause is usually not willpower, and the four reasons rules get broken sorts out which one applies.

This is also the reason a broker export cannot rebuild your R column for you. MetaTrader's history export carries one stop-loss column and it holds the final value, so a widened stop is indistinguishable from a stop that never moved. The original risk exists only where you wrote it down before entering.

3. Losses floored at exactly -1R

A stop is a resting order, not a guarantee. Gaps, news-release slippage and weekend holds all fill worse than the level.

Most journals subtract exactly 1 for every loss, ours included: our cumulative R adds each winner's R and subtracts a flat 1 for each loser. It is the right default for a discretionary intraday trader whose stops mostly fill where they were put, and it is optimistic for anyone holding through NFP or trading thin hours. If a gap costs you 2.4 times your risk, record -2.4R. The stat exists to tell you the truth about risk, so an outcome worse than the plan is precisely what it must not round away.

4. Partial exits counted as the best fill

Close half at 1R and half at 3R and the trade is not a 3R trade. Weight it: half of 1 plus half of 3 is +2.0R.

The same applies to scaling in, where the risk is the total exposure against the stop, not the risk on the first tranche. If you scale, the arithmetic is worth doing once carefully and then automating, because doing it by feel biases in one direction only.

And the case where there is no R at all

If you did not define the risk before entering, there is no denominator. A trade managed by feel has no R-multiple, and the honest log records that rather than back-fitting a stop level that was never there. Mental stops count only if you write the level down first. This is the real reason "log the stop" matters more than any other field: without it, every risk-adjusted number on the page silently stops existing.

What R Still Will Not Tell You

R is a per-trade ratio. It has no opinion about anything that lives between trades.

• Order. +18R arriving as a steady climb and +18R arriving after a 12R drawdown are the same number and different accounts. R has no memory, so pair it with the equity curve.
• Time. 30R over 500 trades and 30R over 50 trades are not the same edge. Divide by the trade count before comparing anything.
• Sample size. A 4R month can be one lucky trade. The expectancy post covers how many trades it takes before an average in R stops moving, and the quick version is to delete your best trade and recompute.
• Correlation. Three concurrent long positions on EUR/USD, GBP/USD and AUD/USD are one trade wearing three hats. Each shows as 1R of risk and together they are closer to 2.5R of exposure.

Where R Lives in TradingSFX

Nothing here needs a spreadsheet, but it does need the stop recorded on every trade.

• The trade form derives realised R for you. Enter the entry, the stop and the exit and the R:R field fills itself from those three prices, magnitude only. Type your own number and it stops overwriting you, which is the escape hatch for a widened stop or a weighted partial exit.
• The dashboard has a P&L / R:R toggle. Flip it and the stat cards switch from dollars to R, so average win becomes average win in R and profit factor becomes cumulative R. Available on the free plan.
• The TradingView indicator pastes the planned figure. Set the trade up on the chart, paste it in, and the R:R that arrives is entry to target over entry to stop, marked as manual so the realised calculation does not overwrite it. That is the planned-versus-realised distinction from section one, made explicit at the point of entry.
• Per symbol and per strategy. The detailed analysis and strategy breakdown pages recompute average R and cumulative R for each slice, which is the comparison this whole article is about. Pro and above.
• Backtest runs are measured the same way. The replay backtester reports net R for the session alongside win rate, so a backtested setup and a live one arrive in the same unit. Replay and practice trades are on every plan, including the free one; saving those trades into the journal is Pro and above.

You can see the R view running on sample data without an account in the detailed analysis demo.

Bottom Line

An R-multiple is one division: result over risk. What makes it useful is that both halves cancel out everything except the quality of the decision, and what makes it dangerous is that it cancels out your sizing too.

So keep three habits. Write the stop down before you enter, or the number does not exist. Log what the trade actually did, not what it was supposed to do, and record a loss worse than the stop at its real size. Then read the R total next to the equity curve, and treat any disagreement between them as the most interesting thing on the page.

Every risk-adjusted stat in this article needs the same three fields underneath: an entry, an exit and a stop. TradingSFX Basic is free forever at 10 trades a month, which is enough to get an R column started and see whether yours agrees with your account.

Published August 28, 2026. All figures in this article are arithmetic worked openly from the stated inputs. No survey data, study results or user statistics are cited, and no competitor pricing or features are quoted. The attribution of the R-multiple term to Van K. Tharp's Trade Your Way to Financial Freedom was checked on 28 August 2026.