Expectancy and R-Multiples: Calculating Average Profit or Loss per Trade

A claim of a 70% win rate does not tell you whether a method makes money. The useful question is how much remains, on average, after one trade. That number is expectancy.

Expectancy: average PnL per trade

Expectancy is the average amount earned or lost per trade when the same rules are repeated many times. It considers how much remains, rather than only how often trades win.

Weight winning and losing outcomes by their probabilities and combine them.

A 40% win rate can have positive expectancy. Losing more often is compatible with a profit when wins are twice as large. The reverse can also occur.

These examples explain why win rate alone cannot select a strategy. An 80% win-rate method can be worse than a 40% one.

R-multiples: a common unit for PnL

Dollar amounts are difficult to compare across different account sizes and assets. Traders therefore express PnL in R.

1R is the amount originally planned to be risked on that trade: entry-to-stop distance multiplied by quantity, or the planned loss amount.

R provides the same measurement scale whether the account is $500 or $50,000 and whether it trades BTC or an altcoin. Setting 1R uses the same calculation as position sizing: choose the risk amount first and work backward to quantity.

Expectancy measured in R

After converting each trade to R, expectancy is also expressed in R, making strategy comparisons easier.

This means an average of 0.2R per trade. If 1R is 2% of account equity, the corresponding simple per-trade expectation is 0.4% of equity.

Positive expectancy is necessary for more trades to improve the expected result. Repeating a negative-expectancy rule more often accelerates expected losses. Check its sign before increasing frequency.

How it differs from profit factor

Both summarize a strategy, but answer different questions.

Profit factor, or PF, is gross profit divided by gross loss: a ratio describing how much was earned per unit lost.
Expectancy is the amount per trade, in money or R.

Two strategies with PF 1.3 can have different expectancies, such as +0.05R and +0.4R per trade. Fees may erase the first while leaving the second positive. The guide emphasizes expectancy when assessing practical tradability.

Expectancy alone misses how difficult the path was. Maximum drawdown, or MDD, helps describe that. Review all three together.

Trap 1: expectancy before costs

Backtest defaults may omit fees and slippage. The smaller the edge, the more consequential that omission becomes.

Apply the same costs to a +0.10R strategy and the result is $10 − $9.60 = +$0.40, effectively near zero. The strategy did not change; the measurement became more complete.

Check which cost basis produced the number. Use the fee calculator with your exchange's actual rate, and read backtesting for why costs can determine viability. Comparing expectancies calculated under different cost assumptions is not meaningful.

Trap 2: changing the stop distorts the R framework

The guide's R framework requires defining initial risk before entry and retaining that reference. Moving a stop farther away can turn a planned −1R loss into −2R or −3R.

The win rate is unchanged, but the sign reverses because actual losses exceeded the original risk unit. The guide stresses following the defined stops and recording actual outcomes against the original R. Losing that discipline can increase risk of ruin.

The opposite problem also matters. Exiting anxiously at +0.3R before the intended target reduces the average winning R. Small wins combined with large losses can produce negative expectancy despite a high win rate.

Trap 3: too few trades

An expectancy from fifteen trades can be dominated by noise. One unusually large winner may inflate it substantially.

The guide proposes at least 100 trades spanning rising, falling and ranging markets. It argues that 120 trades taken only from two bullish months primarily describe one regime rather than a broad sample.

Three levers discussed for improving expectancy

The guide groups potential improvements into three areas.

① Improve win rate: Narrow entry conditions. This can reduce trade count and total profit, and fitting filters only to historical winners risks overfitting.
② Increase average winning R: Let winners run farther, adjusting reward relative to risk. Win rate often falls in exchange.
③ Reduce costs: Lower trading frequency or fee rates. With a small edge, the cost example shows how material this can be.

Win rate and payoff size often trade off against each other. The task is to find a combination with higher expectancy, rather than assume both can increase independently. The Kelly criterion then addresses how much to risk on a trade under its assumptions.

Calculate it from your own records

The procedure has six steps.

1. Collect only fully closed trades, excluding unrealized PnL.
2. Record each trade's risk amount defined at entry. That is its 1R.
3. Deduct fees from realized PnL and divide by 1R.
4. Add all R-multiples and divide by trade count to obtain expectancy in R.
5. Remove the largest winner and recalculate.
6. Split the record into earlier and later halves and calculate each separately.

If the sign survives steps 5 and 6, the guide considers the estimate more credible. The initial risk amount cannot be reliably reconstructed from memory, so record entry, stop and quantity at the time in a trading journal.

Three key points

① Expectancy = Win rate × Average gain − Loss rate × Average loss. Its sign, rather than win rate alone, indicates an expected profit or loss.
② 1R is the loss amount planned before entry. R allows comparisons across account sizes and assets.
③ Costs can reduce +0.2R toward +0.1R, and postponed stops can turn −1R into −3R. Either can reverse expectancy.

Caution

Win rates, PnL amounts, fee rates and trade counts are hypothetical calculation examples, not actual performance of a particular strategy. Historical expectancy does not guarantee future returns and changes when the sample changes. Decisions and their consequences remain your responsibility.