Risk & Psychology · beginner · 7 min

Position Sizing

Ask a room of traders what matters most and most will say picking the right entry. The arithmetic says otherwise. The single biggest determinant of whether an account survives is not which trades you take — it is how much you stake on each one. A brilliant strategy sized carelessly still blows up; a mediocre one sized sensibly can grind on for years. This lesson is about that second thing.

What "size" actually means

Position size is the amount of the account exposed to a single trade's outcome. The useful way to express it is not lots or shares but risk per trade: the fraction of the account you lose if the trade hits its stop. If you have a £10,000 account and you would lose £100 when your stop is hit, you are risking 1%.

Fixing risk as a percentage of the current account, rather than a fixed cash amount, has one important property: as the account shrinks, the cash you risk shrinks with it. You never bet the same £100 into a smaller and smaller pot. That is the mechanism that keeps a bad run from becoming a fatal one.

The arithmetic of a losing streak

Losing streaks are not rare accidents; they are guaranteed. A strategy that wins 50% of the time will, over a few hundred trades, string together runs of six, eight, ten losers somewhere. The only question is whether your sizing lets you sit through them.

Here is what a run of ten straight losers does to an account at different risk levels. Each loss is taken on the reduced balance (percentage-of-account sizing):

Risking 1%: 0.99¹⁰ ≈ 0.904 → about a 10% drawdown. Uncomfortable, survivable.
Risking 2%: 0.98¹⁰ ≈ 0.817 → about an 18% drawdown.
Risking 5%: 0.95¹⁰ ≈ 0.599 → about a 40% drawdown.
Risking 10%: 0.90¹⁰ ≈ 0.349 → about a 65% drawdown.

The last row is the trap. Nothing about the strategy changed between the first row and the last — same win rate, same edge, same ten losers. Only the size changed, and it turned a survivable dip into a hole that is very hard to climb out of. Why "hard" is not an opinion is the subject of Drawdown and Recovery: a 65% loss needs a 186% gain just to get back to even.

Sizing from the stop, not the other way round

There is a common ordering mistake: decide the position size first, then place the stop wherever there is room. That lets the trade dictate your risk. The disciplined order is the reverse:

Decide the fraction of the account this trade may lose — say 1%.
Find where the stop belongs on the chart — the level that says the idea was wrong.
Back out the size from those two. Size = (account × risk%) ÷ (distance to stop).

A wider stop therefore means a smaller position, not more risk. The risk is held constant by construction; the position adjusts to it. This is the opposite of the instinct to "size up because I'm confident," which quietly enlarges risk on exactly the trades where conviction is running the show.

What this does and does not promise

None of this is a claim that small position sizing makes a strategy profitable. A negative edge sized carefully still bleeds out slowly — it just takes longer. Sizing is not a substitute for having something worth trading; it is what keeps you at the table long enough for a real edge, if you have one, to show up in the results. Whether you have an edge is a separate, harder question, and the honest way to answer it is covered in The Overfitting Trap and Reading a Robustness Report.

The register to keep is survival first. The next lesson, Risk of Ruin, makes the point exact: it shows how even a positive edge has a real probability of blowing up when it is sized too large.

Seeing your own sizing

The Trading Journal widget is where sizing stops being theory. Logging risk-per-trade alongside outcomes lets you check whether you actually held risk constant or crept it up after wins and down after losses — a pattern almost nobody believes they do until the log shows it.