R multiple: comparing trades of different sizes, with profit factor
Dollar P&L mixes trade quality with position size. Expressing every trade as an R multiple after costs makes trades comparable and shows when size, not the setup, produced the result.
Start in plain English
A small-cap scalper rarely trades the same size twice. Share count changes with stop distance, spread, liquidity and buying power, so a $300 winner on Tuesday and a $300 winner on Thursday can be very different trades. This lesson shows how to use the R multiple to put every trade on one scale, read the shape of your results, and catch the moment when size, rather than the setup, starts producing your P&L.
The core idea, stated once: an R multiple is a trade's result after costs divided by the risk you planned before entry. If you planned to lose $50 at your initial stop, that $50 is 1R. A trade that nets +$75 is +1.5R; one that nets −$65 is −1.3R. The short definition and its caveats live in the glossary entry R-multiple; this lesson is about using it on a real log.
Two conventions make R comparable across trades, and this course uses both:
- 1R is fixed before entry: (planned entry − initial stop) × planned shares, for a long. Moving the stop later does not change 1R for that trade.
- R is measured after costs: commissions, fees and any slippage come out of the result before you divide.
Three logs with the same +$400
All learners, trades and numbers in this section are Illustrative.
The plan, the same for all three learners. Risk $50 per trade at the initial stop (1R = $50). Target 2R, stop 1R, time stop 10 minutes. 20 trades. Size each trade from stop distance: shares = $50 ÷ per-share risk.
All three finish the block at +$400 net. Here is what their logs look like in R.
Path ASame risk on every trade
Every trade risked $50. 8 trades reached the target (+1.9R each after costs), 7 hit the stop (−1.0R), 5 were scratched (−0.04R).
Sum: 8 × 1.9 − 7 × 1.0 − 5 × 0.04 = +8.0R. Average +0.40R per trade. In this log, dollars and R tell the same story, because size did not change.
Path BSize drifted after losses
17 trades risked $50 and summed to −4.0R (−$200): 4 targets at +1.9R, 11 stops at −1.0R, 2 scratches at −0.3R. After two consecutive losses, three trades were taken at four times the planned size ($200 risk) and returned +2.0R, +0.5R and +0.5R (+$600).
Sum: −4.0R + 3.0R = −1.0R over 20 trades. The dollars say +$400. The R log says the planned-size trades lost, and three oversized trades covered it. Each trade's 1R is the risk fixed at its own entry: $50 for the 17, and $200 for the three oversized trades, whose size and stop were set before entry even though they broke the plan. Measured against the $50 plan instead, those three would read +8.0R, +2.0R and +2.0R, and the R log would hide the size change. Had those three each lost 1R, the block would have ended at −$800.
Path CMostly small results and one outlier
Every trade risked $50. One trade ran to +7.4R. The other 19: 5 at +1.0R, 6 at −1.0R and 8 small scratches at +0.2R.
Sum: +8.0R, identical to Path A. Without the one outlier, the other 19 trades sum to +0.6R, about +0.03R per trade. The median trade is +0.2R.
Same dollars, three different processes. Path A and Path C have the same total but rest on different evidence: one on a repeatable range of results, the other on a single trade. Path B's dollar result came from a size change, not from the plan. The rest of this lesson shows how to compute R from a log, how to read R distributions and profit factor, and how to watch for the drift in Path B.
Computing R from a trade log
For each trade, record these before entry: planned entry (or order ceiling), initial stop, planned shares, and planned risk in dollars. After exit, record the fills, fees and any slippage penalty. Then:
- Gross R = (exit fill − entry fill) × shares ÷ planned risk (sign reversed for shorts).
- Net R = (gross P&L − fees − slippage penalty) ÷ planned risk.
A worked example (Illustrative; Larkspur Mobility is fictional). Planned entry $5.42, initial stop $5.30, 400 shares. 1R = $0.12 × 400 = $48. The entry fills at $5.42 and the exit at $5.61. Fees are $0.005 per share round trip, so $2.00.
Net R = ($0.19 × 400 − $2.00) ÷ $48 = $74.00 ÷ $48 = +1.54R.
Edge cases that change the number:
- The stop was moved. R stays based on the initial stop. If you trail the stop to break-even and get stopped, the result is about 0R minus costs, not "a 1R loss avoided".
- The entry filled worse than planned. Use the planned risk as the denominator and let the worse fill show up as a smaller result. Some traders divide by risk at the actual fill instead; pick one and write it down.
- Partial exits and adds. Each partial is part of one trade's result; divide the trade's total net P&L by the trade's planned risk. Scaling in and out covers the cases where adds change the risk.
- No stop was planned. Then there is no 1R. Do not invent one after the fact; flag the trade as "no planned risk" and count how often that happens.
Reading an R distribution
Once every trade has an R value, look at the whole distribution before looking at the average.
Four readings are worth writing next to any R distribution:
- Mean and median. When they differ a lot, a few trades carry the total (Path C).
- Share of losses beyond −1R. Each one is a stop that filled late, a halt, or a stop that was moved or skipped. In a plan with hard stops, these should be rare; count them.
- Share of the total from the largest trade. In Path C, one trade is 7.4R of an 8.0R total.
- Count. Twenty trades is a small sample; see the section on sample size below.
Profit factor and payoff ratio
Two summary ratios appear in many trading tools. Both can be computed in R or in dollars.
- Profit factor = total won ÷ total lost, after costs. Above 1.0, the block made money; below 1.0, it lost.
- Payoff ratio = average win ÷ average loss.
Both depend on choices that tools rarely agree on. For Path A, profit factor is 15.2 ÷ 7.2 = 2.11 if scratches count as losses, and 15.2 ÷ 7.0 = 2.17 if they are left out (Illustrative, computed).
As of 2026-10-05, on Plus and above the FennEdge Trading Journal shows profit factor and payoff ratio for broker-synced trades, computed from dollar P&L. It classifies a trade as a win or loss by its net P&L after the fees on its own fills, so a gross-flat trade that paid fees counts as a loss, and it computes profit factor as the winners' net P&L divided by the losers' net P&L; a gross win rate is shown for reference only. That matches the first Path A figure above, where scratches that paid fees count as losses. It is one reasonable convention; your spreadsheet may use another. Compare numbers only when you know the convention.
Profit factor in dollars and in R can disagree, and the gap is informative. In Path B, dollar profit factor is $980 ÷ $580 = 1.69; R profit factor is 10.6 ÷ 11.6 = 0.91 (Illustrative, computed). A dollar ratio well above an R ratio means larger sizes landed on the winners. That can be luck or drift, and it is worth checking either way.
Observe vs. infer
Learner B's weekly summary reads: "+$400, profit factor 1.69, the setup works." Separate what the log shows from what is being read into it.
| What the log shows (observed) | What is inferred | Why it is not established |
|---|---|---|
| Net P&L +$400 over 20 trades | "The setup works" | The 3 oversized trades made +$600 and the other 17 lost $200; in R the block lost 1.0R |
| Dollar profit factor 1.69 | "Strong edge" | The R profit factor is 0.91; the dollar ratio reflects size, not trade quality |
| Three trades at $200 risk after two losses | "Pressing an edge" | The size change followed losses, not a written rule; the log shows timing, not reasons |
| 20 trades | "Enough to judge" | A rough 95% range for average R at this standard deviation runs below zero (see below) |
What changes the R reading
R is only as good as the inputs. These conditions change what a given R value means:
- Stop placement. A tighter stop makes 1R smaller in dollars, so the same exit becomes a larger R multiple. Comparing R across setups with very different stop logic needs care; MAE and MFE shows how to check whether stops are being hit by noise.
- Moved or skipped stops. These turn planned −1R losses into −2R or −3R. Track the share of losses beyond −1.2R as its own number.
- Costs. On a 15-cent target, $0.005 per share in fees plus two cents of average slippage takes about 17% of the gross move before you divide (Illustrative).
- Halts. A halt can turn a 1R stop into a −4R fill. That risk belongs in the distribution even if it is rare in your log.
- Size drift. Covered next.
Detecting size drift
Size drift is when actual risk per trade wanders from planned risk, often after wins (confidence) or losses (trying to get it back). Three checks catch most of it:
- Actual risk ÷ planned base risk, per trade. Plot it in trade order. A run of values above 1.0 after losses is the Path B pattern; Revenge trading covers it as a behaviour.
- Dollars vs. R. Multiply your R total by your base risk and compare with your dollar total. Path B: −1.0R × $50 = −$50, against +$400 actual. A large gap in either direction means size varied with outcome.
- Profit factor in R vs. in dollars. The 1.69 vs. 0.91 gap above.
None of these proves a mistake. Planned size changes (for example, smaller size in premarket) are fine if they are written into the plan before the trade.
Cost of being wrong
Using the Larkspur Mobility trade (400 shares, 1R = $48, entry $5.42, fees $2.00; Illustrative):
| Outcome | Exit | R after costs | Dollars after costs |
|---|---|---|---|
| Target | $5.61 | +1.54R | +$74.00 |
| Stop, filled at the stop | $5.30 | −1.04R | −$50.00 |
| Stop, 3 cents of slippage | $5.27 | −1.29R | −$62.00 |
| Halt, reopens at $4.95 | $4.95 | −3.96R | −$190.00 |
The cost of ignoring R is the Path B case: a log that shows +$400 while the planned-size process lost money. Scaling up that process on the strength of the dollar total would put more money behind a record of −0.05R per trade across all 20 trades, and about −0.24R per trade (−4.0R ÷ 17) on the 17 trades taken at the planned size.
How many trades before this means anything
For Path A, average R is +0.40 with a standard deviation of about 1.31R. A rough 95% range for the true average after 20 trades runs from about −0.17R to +0.97R; with the same standard deviation over 100 trades, from about +0.14R to +0.66R (Illustrative, normal approximation). Twenty trades cannot tell this process apart from a break-even one. How many trades before your numbers mean anything explains the method.
Sharpe-style ratios (average return ÷ its standard deviation) have the same problem. On short samples they carry wide estimation error, which Lo (2002) quantifies; a high ratio from 20 trades is weak evidence.
What this lesson does not claim
- It does not claim any R value or profit factor is good or bad in general. The numbers here are Illustrative.
- R does not measure skill by itself. It measures results relative to a plan, so a poorly placed stop produces misleading R values.
- A profit factor above 1.0 in a past block does not indicate future results.
- The Trading Journal convention described above is current as of 2026-10-05 and may change.
Go deeper
- FennEdge first.
- Trading Journal: if you trade live, it syncs real fills from supported brokers; on Plus and above it shows profit factor, payoff ratio, win rate, average winner and loser, maximum drawdown and maximum consecutive losses, computed from dollar P&L rather than R. As of 2026-10-05 it has no planned-stop field, so it cannot compute R; keep R in your own sheet next to it. The Journal holds broker-synced trades only, not practice or simulated entries.
- Journal Review (Pro or Max, as of 2026-10-05) lets you tag trades and mark them "Within plan" or "Outside plan", which pairs well with an R column kept separately.
- DAS course: Build a sizing calculation you can audit shows share count from dollar risk; An order audit and reconciliation worksheet shows how to reconcile fills, which is where net R comes from.
- Glossary: R-multiple for the definition and its caveats.
- Expectancy in R. Average net R per trade is expectancy in R; Expectancy covers the formula and Risk/reward and breakeven win rate covers planned vs. realized R:R.
- R by tag. Grouping R by setup tag turns a log into base rates per setup; Reviewing scalps and Building a playbook use this.
- Sharpe-style ratios. Useful for comparing long return series; on a few dozen scalps they mostly measure noise.
- Next: Backtesting intraday ideas applies R to historical tests; Capstone: a 100-trade edge review puts all of it together.
Exercises
Six exercises follow this lesson: three concept checks (which 1R to use after a stop is moved; dollars vs. R; reading a profit-factor gap) and three calculations (net R for one trade after fees; profit factor in R from a block summary; total R for a block where size changed). Exercises, answers and progress are available after signing in.
Primary sources
- Sharpe, W. F. (1966). Mutual Fund Performance. The Journal of Business 39(1).
- Lo, A. W. (2002). The Statistics of Sharpe Ratios. Financial Analysts Journal 58(4).
- FINRA. Frequent Intraday Trading: Understanding the Basics.
Sources checked 2026-10-05. Larkspur Mobility, the three learners and all trades, sizes and costs in this lesson are fictional and Illustrative. This lesson is education, not a recommendation to buy or sell any security. Past patterns do not indicate future results.