R-Multiples and Risk/Reward in Practice

Why this separates professional trading from guessing

A strategy can look good because it has many winners, yet still lose money if the losers are too large. It can also have a low win rate and be profitable if the winners are big enough. R-multiples force this accounting to the surface in a simple way.

For futures traders, R is especially useful because contracts have very different point values. One oil trade, one Nasdaq trade and one gold trade can't be compared on nominal dollar result alone. They have to be compared against the risk that was actually set before entry.

Traders often talk about a trade offering "three to one" or "at least 1:2." That sounds precise, but says little without answers to three questions: How realistic is the target? How often does the strategy hit? And is the loss actually limited to the amount that was defined?

R-multiples give a common measuring language. An MNQ trade with $150 in planned risk, a gold trade with $400 and an oil trade with $250 can all be assessed on the same scale. That makes it possible to separate the quality of the process from account size and instrument.

What is one R?

R stands for the original planned risk in the trade:

Then add expected commission, fees and a conservative slippage buffer if you want to measure real account risk.

Example with MNQ:

• entry: 20,000
• stop: 19,940
• distance: 60 points
• value: $2 per point
• number: 2 contracts

The price risk is $60\times2\times2=240$ dollars. If the estimated total cost and slippage is $10, practical 1R is roughly $250.

The result is expressed as:

A net result of $500 is then +2R. A loss of $125 is -0.5R.

Why risk must be frozen at entry

If R is redefined after the trade, the metric loses its value. A trader who moves the stop and then calls the larger amount "1R" is hiding a process breach. R should therefore be calculated from the valid plan at entry and kept as the reference in the journal.

You can keep two columns:

• Planned R: risk based on entry, stop and expected costs
• Realized R: actual net result divided by planned R

If a stop order fills badly and the loss becomes -1.3R, that is exactly what the journal should show.

Risk/reward ratio

The relationship between planned gain and planned risk can be written:

A trade with $100 in risk and $300 to the target has a 3:1 gain/risk, often referred to as risk/reward 1:3. The terminology is used inconsistently. So write "the target is 3R" rather than just "ratio 3."

This is a planned relationship. The market doesn't owe you the target, and gaps or slippage can make the loss bigger than 1R.

Expectancy: the number that links win rate and payoff

Expected value per trade can be expressed in R:

where $p_v$ is win probability, $\overline{R_v}$ average winner, $p_t$ loss probability and $\overline{R_t}$ average loss.

Assume:

• win rate: 40 percent
• average winner: +2R
• average loser: -1R

Then:

Over a large number of comparable trades, the expectation is 0.20R per trade before any costs not already included.

A strategy with 70 percent winners but only +0.4R on average, and -1R on losers, has:

A high win rate can therefore hide a negative expected value.

Break-even win rate

If the average loss is 1R and the average gain is $W$ R, the break-even win rate before costs is:

The table explains why trend strategies can be profitable with many losses, and why scalping with small targets requires very high precision. After commission and slippage, actual break-even is higher.

1:3 isn't automatically better than 1:1

A more distant target increases the theoretical gain, but often reduces the probability of reaching it. If a 3R target only hits 15 percent of the time and the rest lose 1R, the expectation is -0.40R. A 1R target with a 60 percent hit rate gives +0.20R.

Risk/reward must therefore be observed in actual results. A chart always lets you draw an attractive target; the history shows whether the market usually reaches it before the stop.

R-multiples and position size

R is an amount, not a universal percentage. First you define acceptable account risk, then stop distance, and finally the number of contracts:

A $30,000 account with 0.5 percent risk has a budget of $150. If the setup requires a 90-point stop in MNQ, one contract is already $180 before costs. The trade doesn't fit the risk framework. The solution isn't a random 70-point stop; it's a smaller instrument, a larger account, or no trade.

Read our separate guide to position sizing as well.

Three complete trade examples

Example 1: full loss

Planned risk is $200. The stop fills as planned, and the net loss is $204 after costs:

The strategy did roughly what it was supposed to. A loss isn't automatically a bad trade.

Example 2: partial gain

Two contracts risk $200 combined. The first contract closes at +1R for half the position; the second at +3R:

The move must be calculated with the correct weighting. It's wrong to add 1R and 3R to make 4R when half was closed at each level.

Example 3: stop moved to breakeven

Half is taken at +1R, the rest is stopped at entry. Before costs:

The "winner" is only +0.5R. If many large potential winners are stopped at zero, the breakeven rule can weaken the strategy's expectancy even though the win rate looks prettier.

Scaling in and out

When a position is built in several parts, the combined risk must be measured against one shared invalidation. Assume:

• the first contract risks $100
• the next planned addition risks $80
• the third addition risks $70

Full planned risk is $250, not $100. If additions are only made after the first stop has been raised, the journal should show both the original and the maximum simultaneous risk.

On a partial exit, weighted R is calculated. This prevents the journal from overrating trades where a small remainder hit a large target.

MAE and MFE: what happened between entry and exit?

Maximum Adverse Excursion (MAE) is the largest unrealized move against the position, measured in R. Maximum Favorable Excursion (MFE) is the largest move in favor.

If a trade closes at +0.5R but was +2.5R at its best, the exit rules may be too slow. If winners regularly have an MAE of -0.8R before turning, a stop at -0.5R is probably too tight. But don't optimize the stop on the same dataset the strategy is being assessed on; that increases the risk of overfitting.

A useful journal includes:

Costs and slippage in R

If 1R is $50 and the round-trip cost is $5, the trade in practice starts at -0.10R. If 1R is $500, the same cost is -0.01R. Small risk amounts and frequent trading are therefore more sensitive to friction.

Measure net-R after:

• commission and exchange fees
• bid/ask spread
• normal slippage
• extra slippage around news
• currency effect if the account is denominated in another currency

Gap risk: losses can exceed 1R

A stop order is a trigger, not a price guarantee. Overnight gaps, limit moves or a sudden dry-up in liquidity can produce -1.5R, -3R or more. The strategy's risk analysis should therefore distinguish between:

• planned risk: stop price under normal conditions
• expected realized risk: including typical slippage
• tail risk: rare but severe scenario

For swing trading and commodities, stress-R should be used in the portfolio calculation.

Drawdown in R

Because R normalizes size, drawdown can be compared across accounts. Ten losses in a row is -10R if each loss is full and the risk is held constant.

The probability of losing streaks is often underestimated. With a 40 percent win rate, the loss probability is 60 percent. The simplified probability of five losses in a specific sequence is $0.6^5\approx7.8$ percent. Over a hundred trades there are many possible starting points, so the probability of at least one such streak is far higher.

Size 1R so that a statistically normal losing streak doesn't lead to panic, rule-breaking or forced downscaling.

Fixed percentage or fixed amount?

With a fixed percentage, dollar-R falls after losses and rises after gains. That protects capital in a drawdown, but makes historical dollar results less directly comparable.

With a fixed amount, the process is simple and results are easy to understand, but the same R becomes a larger share of a shrinking account. A hybrid can use a fixed amount within defined account bands.

The most important thing is avoiding R being raised impulsively after losses or reduced randomly out of fear.

Total portfolio risk

Three open trades with 1R risk each doesn't necessarily mean just three independent R's. Long NQ, long ES and short government bonds can all lose on the same inflation surprise.

Total portfolio risk is the sum of planned open risk, adjusted for correlation and gap scenarios. Set an upper limit for simultaneous exposure and group trades by common driver.

Profit factor, expectancy and R

Profit factor is gross gain divided by gross loss. A profit factor above 1 indicates a historically positive gross result. Expectancy in R tells you how much the strategy earned per trade on average.

Neither should be used alone. Also consider:

• number of trades
• uncertainty around the estimate
• largest drawdown in R
• stability between periods
• result per setup and regime
• share of gain from the few best trades

A strategy with +0.25R expectancy over 20 trades is far less documented than the same number over 500 independent, comparable trades.

How to build an R-based trading plan

1. Define technical or fundamental invalidation.
2. Measure stop distance and the contract's dollar value.
3. Decide the maximum risk amount before position size.
4. Round the number of contracts down.
5. Record the planned target in R, but only allow rule-based exits.
6. Include costs and realistic slippage.
7. Set maximum daily, weekly and total portfolio risk in R.
8. Evaluate expectancy, MAE, MFE and drawdown after a meaningful series.

More models can be found in the overview of futures strategies.

Common mistakes

Calling the target distance the expected gain. A 4R target is worthless if it almost never hits. Moving the stop and redefining 1R. That hides a risk breach. Ignoring partial size. Exits must be weighted. Counting breakeven as a full winner. The result is around 0R before costs. Comparing gross-R. Costs can flip small positive systems. Optimizing on too few trades. High expectancy can be caused by a handful of outliers. Believing 1R is the maximum possible loss. Gaps and slippage can exceed the stop.

Conclusion

R-multiples make risk concrete and results comparable. They force the trader to define the loss before the gain and show why win rate or an attractive price target alone don't prove profitability.

But R is a measurement tool, not a strategy. The real quality lies in positive expectancy after costs, acceptable drawdown and consistent execution across different market regimes. When the journal also contains MAE, MFE and process errors, "risk/reward" becomes a decision system - not a slogan.

Capital risk: Futures trading carries a high level of risk and is not suitable for everyone. Losses can exceed the capital deposited. This article is general information and not investment advice.

Sources and further reading

• CME Group Education
• CFTC: Futures Market Basics
• NFA: Investor Resources

Frequently asked questions about R-multiples

Short answers about 1R, risk/reward, expectancy, partial exits, slippage and how R is used in the journal.