Exits, partials and R-multiples: how to measure a trade without fooling yourself

In the beginner course you met the reward-to-risk ratio. Here we go further with the tool professionals use: the R-multiple. It's the simplest and most honest way to measure every trade, whatever the account size or the instrument.

What R is

1R is the loss you take if your stop is hit. If you risk $100 per trade, 1R = $100. Every result is measured in R:

• stop hit: −1R
• closed at +$200 with $100 risked: +2R
• closed at breakeven: 0R
• closed with a $50 loss: −0.5R

The big advantage is comparability: a +2R trade on a $1,000 account and one on a $100,000 account are the same result. And managing trades turns into simple arithmetic: the sum of your R tells you how the system's doing, regardless of size.

Expectancy in R

Expectancy is the average result per trade, in R:

Expectancy = (win probability × average R of wins) − (loss probability × average R of losses)

With 40% winners at +2R and 60% losers at −1R: 0.4 × 2 − 0.6 × 1 = +0.2R per trade. Over 100 trades, on average +20R. With 50% winners at +1R: 0.5 − 0.5 = 0: no edge (and costs make it negative). Win rate on its own tells you nothing: it only matters in combination with the ratio of wins to losses.

One take profit, or partial closes?

With a single target, every trade either hits the target or hits the stop (or goes to breakeven). With partial closes you take some of the position off at a first target and let the rest run towards a second. The most common setup: close 50% at +1R, move the stop to breakeven, and leave the other half heading for a more distant target.

The numbers are invented to explain the mechanism: in real life you estimate them from your own data. The point is that choosing between a single target and partials changes the shape of your results (more small wins, fewer long losing streaks), and you judge it by expectancy, not by "feels better". Partials don't magically improve a system: they redistribute the result and usually reduce variability, which helps psychologically.

Breakeven: protection or a choke?

Moving your stop to breakeven after an initial move in your favour removes any chance of losing on that trade. But it has a cost: price often pulls back toward your entry before moving on, and a breakeven set too early will knock you out at zero just before the move you were waiting for. There's no universal rule: what matters is deciding it in advance, writing it down, and measuring in your journal whether, for your system, breakeven improves or hurts expectancy.

Emotional exits

Closing a winner "because there's something in it already" or holding a loser "because it'll come back" are emotional decisions, not system ones. A good exit is one of these three:

1. The planned target has been reached.
2. The stop has been hit.
3. A written invalidation condition in your plan has occurred (for example, the structure that got you in has broken).

Everything else is emotional noise. If you often close before the target or move the target, log it in your journal with the reason: you'll find out what it costs you.

Sample size matters

An expectancy calculated over 10 trades tells you nothing: the statistical error is huge. You need dozens, ideally hundreds, of trades under fixed rules to say whether a system works. A single trade, win or lose, proves nothing about the system or about you.

In short

• 1R is your initial risk: every result is measured in multiples of R, whatever the account size.
• Expectancy in R is the average result: that's the number that matters, not the win rate.
• Partial closes and breakeven change the shape of your results: decide them in advance and measure them.
• Exit on target, stop or written invalidation, not on emotion.

Practice

1. Take your last 30 trades (demo is fine), express each in R and work out the expectancy.
2. Write down your exit rule: target, partials, when you move the stop.
3. For your next 20 trades, note each time whether you followed the rule and what the result would have been had you followed it to the letter.