Brokers
Gold Contract Specifications: 7 Lines That Set Your Costs
Gold contract specifications, line by line: contract size, digits, spread, swap, margin and hours, with each one turned into dollars per trade.
9 min read
Read more →Lesson 03 of 8 · Risk and Money Management
Risk-Reward and Thinking in R
20 min4 topics
Topic 1 of 4
By the end of this lesson
- Express a result in R
- Relate required win rate to a target R multiple
- Spot a plan whose numbers cannot work
Before this lesson
Two traders compare results. One made $340 on gold, the other $85 on EUR/USD. Which trade was better? The question is unanswerable, because dollars measure position size as much as skill. R fixes that.
R as a unit
R is the amount you risked on a trade. Every result is then quoted as a multiple of it.
R multiple = result / amount risked
What each variable means:
- result — Profit or loss in money, after costs
- amount risked — Entry to stop, at the size you traded
A trade that risked $50 and made $150 is +3R. One that risked $50 and lost the stop is −1R. One closed early for $20 is +0.4R.
Back to the two traders:
| Risked | Result | In R | |
|---|---|---|---|
| Gold trade | $340 | +$340 | +1.0R |
| EUR/USD trade | $25 | +$85 | +3.4R |
The smaller trade was much better executed. In dollars it looked like a quarter of the other one, which is exactly the illusion R removes.
Good to know
R also makes your own history comparable to itself. A journal in dollars mixes the year you traded 0.02 lots with the year you traded 0.5; a journal in R shows whether the method improved.
The planned R and the realised R
Before the trade you have a planned R multiple: target distance divided by stop distance. Afterwards you have a realised one. They differ whenever you exit early, take partial profits, or get slipped.
Record both. A persistent gap between them is one of the most useful things a journal can show — it usually means the plan is fine and the exits are not.
Try it now
The profit and loss calculator gives the money figure for a given entry, exit and size. Divide it by what you risked and you have the R.
Win rate and R together
Neither number means anything alone. A 90% win rate is worthless if the tenth trade gives back all nine; a 25% win rate is excellent if the winners are large enough.
Expectancy in R = (win rate x average win in R) - (loss rate x 1)
What each variable means:
- win rate — Fraction of trades that win, as a decimal
- average win in R — Average winning trade, in units of risk
- loss rate x 1 — Losers cost 1R each, by construction
| Win rate | Average win | Expectancy per trade |
|---|---|---|
| 40% | 2.0R | +0.20R |
| 60% | 1.0R | +0.20R |
| 35% | 3.0R | +0.40R |
| 70% | 0.5R | +0.05R |
| 45% | 1.2R | −0.01R |
The first two rows earn identically and feel nothing alike. The 40% system loses six trades out of ten and requires you to keep taking the seventh; the 60% system is comfortable and pays the same. Which one you can actually execute is a real consideration, not a soft one.
The last row is the trap this lesson exists to expose: a respectable-sounding win rate, a positive-sounding reward, and a system that loses money slowly.
The break-even line
For any target R multiple there is a win rate below which the system loses. It is worth knowing by heart.
Break-even win rate = 1 / (1 + R)
What each variable means:
- R — The reward-to-risk multiple you actually achieve
| Reward:risk | Break-even win rate | You need better than |
|---|---|---|
| 0.5 : 1 | 66.7% | Two out of three |
| 1 : 1 | 50.0% | One in two |
| 1.5 : 1 | 40.0% | Two in five |
| 2 : 1 | 33.3% | One in three |
| 3 : 1 | 25.0% | One in four |
| 5 : 1 | 16.7% | One in six |
Read the table as a constraint rather than a menu. A 5:1 system asking for one winner in six sounds easy until you sit through the eleven-loss stretches it produces, and hitting a five-times-risk target requires a market that actually travels that far.
Caution
Use your realised average R here, not your planned one. Planned 3:1 systems routinely realise 1.5:1 once early exits and slippage are counted, and the break-even win rate for 1.5:1 is 40% rather than 25% — a gap wide enough to turn a profitable plan into a losing one.
Plans that fail on arithmetic alone
Some strategies can be rejected before they are ever tested, because their own stated numbers do not work. Run every plan through these three checks first.
- Does the win rate clear the break-even line for the realised R? A scalping plan targeting 0.5R needs to win more than two trades in three, every month, including spread. Very few do.
- Do costs fit inside the target? A 10-pip target on a pair costing 1.5 pips round turn hands 15% of every winner to the broker, and the break-even win rate moves accordingly.
- Is the claimed combination internally consistent? "80% win rate with 3:1 reward" implies an expectancy of +2.2R per trade. At 1% risk that is roughly doubling the account every 35 trades, forever. The claim refutes itself.
A worked rejection
A plan states: 10-pip target, 10-pip stop, "about 55% winners", EUR/USD, one standard lot, 1.2-pip spread.
| Gross win | 10 pips = $100 |
| Gross loss | 10 pips = $100 |
| Spread, both sides | 1.2 pips = $12 |
| Net win | $88 |
| Net loss | $112 |
| Realised reward:risk | 0.79 : 1 |
| Break-even win rate | 56.0% |
| Claimed win rate | 55% |
Nothing about the market was needed to reach that conclusion. The plan is a losing plan on its own figures, and this check took two minutes — which is the argument for running it before the three months of live trading that would have produced the same answer more expensively.
Key takeaways
- R is the amount risked; every result quoted as a multiple of it becomes comparable across pairs, sizes and years.
- Win rate and average R only mean something together — expectancy = (win rate x avg win in R) − loss rate.
- Break-even win rate is 1/(1+R). Compute it from your realised R, not your planned one.
- Costs come out of the target. A plan whose own numbers fail this arithmetic can be rejected before it is tested.