R-Multiples Explained for Crypto Traders

Raw PnL is the number every trader tracks. It's also the number least likely to tell you whether you have an edge.

A $900 winner feels good. But if you risked $1,800 on that trade, it was a lousy outcome: you made 0.5R. That same $900 win on a $300 risk was a 3R winner. Same dollar amount, completely different quality of trade. Without normalizing for risk, your PnL statement measures nothing.

That's what R-multiples solve.

What is 1R?

1R is the dollar amount you agreed to risk before entering the trade. It's simple arithmetic:

1R = (entry price − stop loss price) × position size

If you buy BTC at $104,000 with a stop at $103,000 and size of 0.1 BTC, your 1R is $100. Whether you make $500 or $50 on exit, you express the result as a multiple of that $100 initial risk.

• Exit at $105,000: profit $100 = 1R winner
• Exit at $106,500: profit $250 = 2.5R winner
• Stop triggered at $103,000: loss $100 = −1R loser
• Moved stop down, took a larger loss of $180: −1.8R (and a mistake log entry)

1R must be set at entry, before the trade develops. Changing your stop mid-trade changes the denominator, which is how traders accidentally hide their actual risk behavior from their own journals.

Why PnL lies but R doesn't

Position sizing creates noise in raw dollar PnL that makes comparing trades across sessions impossible.

Say you have two trades in the same week. Trade A: you size up aggressively, risk $2,000, catch a clean move, make $3,000. Trade B: you take a conservative position, risk $400, the setup plays out well, you pocket $1,200.

Dollar PnL says Trade A was more than twice as good. R-multiples say Trade A was 1.5R and Trade B was 3R. Trade B was actually better trading: cleaner execution, better reward relative to risk. Raw PnL masked that.

This matters because traders tend to size up on lower-quality setups (chasing larger gains) and size down on high-confidence setups (protecting themselves when they feel uncertain). These sizing biases mean dollar PnL often inverts your actual skill ranking. R-multiples strip out the size and show what you actually did.

The other way PnL misleads: a single outsized winner can put your month in the green while hiding that you lost money on 18 of 22 trades. If that one winner was 8R, you have a real edge even with a 27% win rate. But if that winner was the result of a lucky, oversized trade you took by accident, the win rate problem is the real story. R-multiples let you see which.

Computing R correctly for crypto

On stocks, R is close to trivial: one fill, dollar-denominated fees, no carry cost. On crypto it takes more care.

Use VWAP, not the first fill. If your entry order filled across five executions at slightly different prices, your real entry price is the volume-weighted average. Using the first fill overstates or understates the true entry depending on price direction during the fill. Your realized R changes accordingly.

Subtract fees from realized PnL before computing R. On most crypto venues, taker fees run roughly 0.05% per side; see your exchange's fee schedule for the exact rate (Deribit is one example where fees vary by contract type). On a $100,000 notional position, taker fees at entry and exit cost roughly $100 total. That's 1R on a $100-risk trade, so fees that size eat the whole risk unit. If you compute R on gross PnL, you're inflating every winner and masking the fee drag.

Include funding in multi-day holds. Perpetual swap funding payments are part of the trade's economics. A long BTC perp held for two days at elevated funding (+0.03% per 8h) pays 6 funding periods × 0.03% × notional = 0.18% of notional out of pocket. On a $50,000 position that's $90. Whether that's 0.5R or 2R depends on your stop distance, but ignoring it means your reported R is wrong.

Worked example with real numbers:

You're long 0.5 BTC perp on Binance at VWAP entry of $102,800. Stop at $101,800. Exit VWAP $104,100.

1R = ($102,800 − $101,800) × 0.5 BTC = $500

Gross PnL = ($104,100 − $102,800) × 0.5 = $650

Taker fees (0.05% each way):
  Entry: $102,800 × 0.5 × 0.0005 = $25.70
  Exit:  $104,100 × 0.5 × 0.0005 = $26.03
  Total fees: $51.73

Funding (held 16h, 2 funding periods, rate +0.01%):
  $102,800 × 0.5 × 0.0001 × 2 = $10.28

Net PnL = $650 − $51.73 − $10.28 = $587.99

Realized R = $587.99 / $500 = 1.18R

Gross R looked like 1.30R. After fees and funding, it's 1.18R. That's not a catastrophic difference on a short hold, but it compounds: a strategy you think is running at 1.5R average might actually run at 1.1R after proper accounting. Over 100 trades, that gap matters.

What good R distributions look like

A strategy's edge lives in its expectancy:

Expectancy = (win rate × average win in R) − (loss rate × average loss in R)

Worked example with real numbers: You've traded 40 setups.

• 17 winners, average win 2.1R
• 23 losers, average loss 1.0R (you take full stops; no premature exits)

Win rate  = 17 / 40 = 42.5%
Loss rate = 23 / 40 = 57.5%

Expectancy = (0.425 × 2.1) − (0.575 × 1.0)
           = 0.8925 − 0.575
           = +0.32R per trade

Positive expectancy at 0.32R means for every dollar risked, you make 32 cents on average. Over 200 trades at $500 average risk, that's $32,000. The number seems small per trade. That's the point. Edge compounds.

What does a healthy R distribution actually look like in a journal?

• Most losses cluster tightly around −1R (you're taking planned stops, not letting losers run)
• Winners are distributed with a long right tail (some 3R–5R trades, not all just barely 1R)
• No cluster at −2R, −3R (those are stop-moves, the expensive habit)
• The biggest losers are never the biggest winners. If your worst trade is larger in absolute terms than your best trade, that signals a sizing or stop-discipline problem

If you see many winners clustered at 0.4R–0.8R, you're probably taking profit too early. If you see a long left tail of −2R to −4R losers, you're moving stops and holding losers. Both patterns are fixable, but you can only see them if you're tracking R.

How a journal automates this

Tracking R manually is theoretically possible and practically almost never done. You need to log your planned stop loss at entry, then pull it back up when you close. Any tool that doesn't persist your planned stop loss through the trade's lifecycle can't compute R. It only has entry and exit prices, which give you gross PnL but not your actual risk relationship.

Viktury stores your planned stop loss and risk amount at the time you open or record the trade. When the round-trip closes, the server computes realized R from your actual VWAP entry, VWAP exit, fees, and funding, not from the price difference alone. The number in your analytics is the real one.

For existing trades where you didn't log a stop, you can backfill the planned risk amount directly. The system uses that as your 1R denominator.

You can see your full R distribution in the analytics dashboard: per setup, per time period, per exchange. That distribution is the most honest picture of your edge you'll have.

Read more about the foundational journaling habits that make R tracking useful in How to Journal Crypto Trades.

Start your free trial at Viktury and connect your exchange accounts in about five minutes.