Position sizing and risk per trade
The one calculation to run before every position. It converts a stop distance into a size, rather than picking a size and hoping.
12 min read · beginner · lesson 6 of 6
What this covers
- Run the position sizing calculation from risk, stop distance and unit value
- Explain why risk per trade is set as a percentage rather than a fixed amount
- Account for correlated positions and total account exposure
- Express results in R so outcomes are comparable across instruments
- Adjust size for an instrument's volatility rather than widening the stop
The calculation
Fix risk per trade as a percentage of the account
A common range is 0.5 to 2 percent. The number is a policy decision made once, away from any open position, and applied uniformly.
Convert to a currency amount
Account balance multiplied by the percentage. On a 10,000 account at 1 percent, the amount is 100.
Measure the stop distance
The distance from entry to the level that voids the position, in pips, points or ticks. This comes from structure and is fixed before size is considered.
Divide
Position size equals risk amount divided by (stop distance multiplied by value per unit). This produces the size at which the stop being reached costs exactly the amount you accepted.
Worked example
IllustrativeAll figures are illustrative.
Account 10,000. Risk per trade 1 percent, so 100.
Entry 1.0900, stop 1.0860. Stop distance is 40 pips.
On a standard lot the pip value is roughly 10 per pip, so the risk per lot is 40 multiplied by 10, or 400.
Position size is 100 divided by 400, which is 0.25 lots.
Run the same calculation with a 20-pip stop and the size doubles to 0.5 lots. The loss if the stop is reached remains 100 in both cases. Size adjusts so the risk stays constant while the structure varies.
Why a percentage rather than a fixed amount
A percentage scales with the account in both directions. After a losing run the amount at risk falls automatically, which slows the rate of loss when performance is worst. After a winning run it rises, which compounds without requiring a decision.
A fixed currency amount does the opposite. It becomes a larger share of a shrinking account, so the sizing grows more aggressive precisely when results say it should not.
The automatic quality is the point. A rule that adjusts without a decision is a rule that still works on the day judgement is worst, which is the day it matters.
Measuring outcomes in R
Once every position risks the same percentage, results become comparable regardless of instrument. One unit of risk is one R. A position that made twice what it risked is a 2R gain; one that hit its stop is a 1R loss.
This removes instrument size from the record. A gain on a large index contract and a gain on a small currency position are directly comparable in R, and a set of results in R describes a method rather than describing which instruments happened to be traded.
It also makes the arithmetic of a method visible. Average R multiplied by the number of positions is the expected result, and improving either the average or the strike rate is a concrete target rather than a vague intention.
Volatility changes the size, not the stop
The same 40-pip stop is generous on one instrument and inside ordinary noise on another. Instruments differ in how far they routinely travel, and so do the same instruments in different conditions.
The correct adjustment is to the size. The invalidation level comes from structure, and the structure on a volatile instrument sits further from entry, which means a smaller position for the same risk amount.
The incorrect adjustment is to hold size constant and accept more risk, or to tighten the stop to fit the desired size. The first breaks the risk policy; the second puts the exit somewhere the analysis says nothing about.
Correlated positions and total exposure
Risk per trade controls one position. It does not control the account when several positions move together.
Long EUR/USD, long GBP/USD and short USD/CHF is three positions with one exposure: short the US dollar. Sized at 1 percent each, the account carries roughly 3 percent on a single outcome. A dollar rally hits all three at once.
Set a second limit on total exposure to any one driver, and count correlated positions against it as though they were one. A third limit on total open risk across everything is worth having as well, since a portfolio of uncorrelated positions can still add up to more than intended.
Risk against reward
Stop distance and target distance together give the ratio on a position. Risking 40 pips to make 80 is 1:2.
The ratio and the strike rate are one calculation, not two. At 1:2, a third of positions reaching target covers the cost of the rest. At 1:1, more than half must reach target. Neither ratio is better in isolation. A ratio is only meaningful next to the rate at which those targets are actually reached, measured over a real sample rather than assumed.
The trap is choosing a ratio that requires a strike rate the method has never produced. A 1:5 target looks efficient and needs only one position in six to work, but only if one in six actually travels that far. Where targets sit has to be justified by what the instrument does, not by what makes the arithmetic comfortable.
Key takeaways
- Size is the output of the calculation. Stop distance is the input.
- Risk per trade as a percentage shrinks exposure automatically during a losing run.
- Measuring in R makes results comparable across instruments and sizes.
- Volatility changes the position size, never the stop or the risk policy.
- Correlated positions are one exposure and must be counted as one.
- A risk-reward ratio means nothing without the strike rate measured beside it.
Common mistakes
- Choosing a size first, then placing the stop wherever that size makes affordable.
- Applying risk per trade correctly to each position while holding five versions of the same exposure.
- Raising risk per trade after losses to recover faster.
- Setting a target ratio that requires a strike rate the method has never produced.
Knowledge check
OptionalRelated
- Leverage and marginMarket Foundations
- Orders and executionMarket Foundations