Position Sizing: The Size Is an Output
Position sizing converts a risk decision into a quantity: the amount you are willing to lose, divided by the distance to the stop, divided by what one unit of movement is worth. The size is the last number in that sequence, never the first.
How it works
It is the step that turns a decision about money into a number of units. You have decided what a losing trade may cost. Position sizing is the arithmetic that converts that into shares, lots or contracts.
The order is the whole discipline. Decide the amount you are willing to lose. Decide where the idea is wrong, from the chart or from a volatility measure. Then divide. The size falls out; it is never chosen.
Shares are the simple case. The amount at risk, divided by the distance in price between entry and stop, gives the number of shares. Every other instrument is this same calculation with one extra conversion in the middle.
In foreign exchange the stop is measured in pips, so you divide the risk by the stop in pips and then by what one pip is worth for the lot size you are using.
In futures the same slot is filled by the tick value. The stop measured in ticks, multiplied by what a tick is worth, is the loss per contract — and the exchange publishes that figure.
Where the formula stops working
A long option cannot lose more than it cost, so the premium paid IS the risk and there is nothing to divide. Sizing becomes a question of how much premium to spend rather than where to put a stop.
A short option inverts the problem entirely. The maximum loss may be undefined, so the sizing question is what a bad outcome would cost rather than what a stop would cost — and those are different numbers.
A fixed size is a moving risk. Suppose a stock’s average true range is 1% of its price in a quiet month and 3% in a busy one. The same number of shares is then three times the bet in one regime that it is in the other.
In practice
Everything scales with the size, including the errors. A fat-fingered entry, a missed exit and a slipped fill all cost in proportion, which is a reason to be conservative while a process is new.
Liquidity sets a ceiling the arithmetic does not know about. A size the formula permits may be more than the market will fill at a sensible price, and that limit binds first in thin instruments.
Horizon and size are linked through the stop. A position held for weeks needs room to breathe, and room means a wider stop, which means fewer units for the same risk.
An opening gap is the case the formula cannot cover. The stop does not fill at its level, so the only thing standing between you and a larger loss is how many units you hold.
They are one decision, not two. Widening the stop after entry without reducing the position silently increases the risk you already accepted.
Cost is proportional too. A round trip costing a hypothetical 0.1% of the position costs twice as many dollars on a position twice the size, so an oversized trade pays more to be wrong as well as losing more.
Doing it in ten seconds
Write three numbers before you look at the order ticket. The money you will lose if wrong. The price at which you are wrong. The value of one unit of movement in the instrument you are trading.
Then divide twice and round down. Risk, divided by the distance to the stop, divided by the unit value. Rounding down rather than up is the whole margin of safety in the procedure, and it costs almost nothing.
Do it every time rather than once. The third number rarely changes; the second changes daily with volatility, and that is precisely why yesterday’s size is the wrong size today. A position carried over from a calmer week is a decision you did not make.
What position sizing is not
It is not a view. Conviction is not a unit.
It is not the same as leverage. Leverage sets the ceiling.
It is not fixed. It moves with volatility.
And it is not optional on a small account. It matters more there.
How this connects
Size is what turns expectancy into survival. A method with a genuine edge can still be ruined by oversized positions, and the risk of ruin rises sharply as the fraction risked on each trade grows.
A losing streak is the stress test. A fixed fraction applied through a run of losses shrinks each position as the account falls, which keeps the drawdown recoverable; a fixed dollar amount does not.
Win rate changes what a streak looks like, not the formula. A method that loses more often than it wins needs a smaller fraction to tolerate the runs it will inevitably produce.
Adding to winners is sizing too. Pyramiding and the anti-martingale approach raise exposure only as a trade moves in your favor, and each addition still has to respect the original risk.
Slippage belongs in the distance. A stop order in a fast market fills beyond its level, so sizing from the exact stop price understates the loss that actually arrives.
Small accounts have finer tools now. Fractional shares and divisible crypto quantities remove much of the rounding problem that whole units create.
Emotion undoes it faster than arithmetic does. Revenge trading and overtrading usually begin with a size the trading plan never allowed, which a daily loss limit and a trading journal are both designed to catch.
And prop firm evaluations make it explicit. Their loss limits translate directly into a maximum size per trade, whatever the chart looks like.
When it fails
In a quiet trading range an oversized position is comfortable, which is how it survives long enough to be dangerous. Nothing signals the error until conditions change.
The second failure is sizing from the balance rather than the stop. A fixed number of shares makes the loss whatever the chart decides.
A third is widening the stop and keeping the size. Both halves of one decision, and only one moved.
A fourth is ignoring correlation. Five positions at correct individual size can be one bet at five times the risk.
A fifth is scaling up after a winning run. The method did not improve; the sample got longer.
And a sixth is rounding up to a whole unit on a small account. Granularity can force a size the arithmetic never permitted, which is an argument for smaller contracts rather than a bigger stop.
The original data
Of the 24,971 videos in the site’s search study, 195 have “position siz” in the title, at a median of 1,738 views across 162 channels — while 14 titles containing “lot size” have a median of 127,651.
The question: Same setup, twice the volatility. Same size?
The professional phrasing has fourteen times the supply and a seventy-third of the audience per video. It is the same calculation under two names, and the name beginners actually type is the one almost nobody publishes under. That gap is the clearest evidence in this corpus that the material is written for the writer.
The answer to that final question is no, and the arithmetic says by how much. If the average true range has doubled, an unchanged size is twice the intended bet, and in a market that has gone from quiet to busy the multiple can be larger still. Recompute the size from today’s volatility before every entry — it takes ten seconds and it is the only step here that cannot be skipped.
Related
Risk per trade is the first number in the sequence and the only one you set outright. Lot size is the same arithmetic expressed in foreign-exchange units. And risk management is the wider frame this single calculation sits inside.
I do not trade the same size every day and call it consistency. Size is a number that needs re-deriving, not one you pick early and leave alone because checking it each morning feels like admin next to hunting the next setup. On the days the market is moving hardest, which are exactly the days it matters, an ordinary size is quietly the biggest bet on the book.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.