How much to put into a position usually gets settled with a fixed percentage of the balance: ten per cent of capital on every trade. That is simple, and it ignores something: two positions opened at the same percentage can risk wildly different amounts of money.
Two separate questions
Two questions get conflated here and they need to stay apart.
How much money am I committing? The size of the position, units times price.
How much money can I lose? That is not position size. It is the distance between your entry and your exit point.
Of two positions with the same money committed, one closing one per cent away and one closing ten per cent away carry ten times different risk. A size worked out as a percentage of capital cannot see that difference.
The three inputs
Three numbers set position size:
- The amount you accept losing. Written as a percentage of capital, this is what you are willing to lose on one trade.
- The distance from entry to exit. As a percentage or as a price difference.
- Capital. The number that turns the first item into an amount.
The division is simple: amount at risk, divided by loss per unit. The result is how many units to buy.
Concretely. Capital 100,000, one per cent risk per trade, so 1,000. Entry at 100, exit at 95, so 5 per unit. A thousand divided by five is two hundred units. The position is 20,000, a fifth of capital.
Run the same arithmetic with an exit at 98 and you get five hundred units, a 50,000 position. Two very differently sized trades, both risking the same 1,000.
What a shorter distance does
The part that surprises people: as the exit moves closer to the entry, the position gets bigger.
Intuition says otherwise. A tight stop feels like less risk, when in fact carrying the same risk requires more units. A strategy built on tight distances ties up a large share of the balance in one position, and a second limit has to come in here.
So the arithmetic needs a cap: if the computed size exceeds some share of capital, the position is trimmed to that share. Trimming means the amount at risk falls below target, which is acceptable. Committing nearly the whole balance to one trade is not.
Changing two numbers at once
The next trap is tuning risk with both the percentage and the distance.
After a losing streak the natural reflex is to do two things at once: cut risk from one per cent to a half, and pull the exit closer at the same time. Together they push size in an unexpected direction and make it impossible to measure which change did what.
The rule: change one number at a time. Adjusting risk, hold the distance fixed; adjusting the distance, hold risk fixed. Otherwise two weeks later you have a result, good or bad, and no record of what caused it.
Positions in the same direction are one position
One last point, at portfolio level. Three positions opened the same way in three assets that move together are not three separate risks.
Sizing per trade without adding it up across the portfolio means “I risk one per cent per trade” actually reads as three per cent. Total risk across correlated positions needs its own cap, and that cap should not be thought of as a multiple of the single-trade limit.
