Position sizing explained

Ask a losing trader what went wrong and you will hear about strategy: the wrong indicator, the fake breakout, the news. Ask a surviving trader and you will hear a number: how much they were willing to lose on any single trade. That number is position sizing, and it is the only input in trading you fully control.

You cannot control whether the next trade wins. You cannot control the spread, the slippage or the news. You control exactly one thing — how much is at risk before you click the button. Get that one thing right and losing streaks become survivable. Get it wrong and no strategy on earth saves the account.

The 1–2% rule

The working convention among professional risk managers is simple: never risk more than 1–2% of the account on a single trade. Not 1–2% of the account in the trade — 1–2% of the account lost if the stop is hit. On a $5,000 account, one percent is $50. Every position you take is built backwards from that fifty dollars, never forwards from “how much can I buy?”.

Why so small? Because losing streaks are not a possibility, they are a statistical certainty. A strategy with a genuine 50% win rate will still hand you eight losses in a row sooner or later — the streak is inside the maths, not a sign the strategy broke. At 1% risk, eight straight losses is a 7.7% drawdown: painful, survivable. At 20% risk it is an 83% drawdown, which needs a 578% gain just to get back to even. That is not a recovery plan; it is an obituary.

The formula

Position size is one division:

Position size = (account × risk %) ÷ stop distance

Everything else is unit conversion — which is what the position size calculator is for. The work worth doing by hand is deciding the two honest inputs: the account figure (what you actually have, not what you wish you had) and the stop (where the trade idea is objectively wrong, not where it “feels safe”).

Worked example — forex

Account: $5,000. Risk: 1% = $50. Trade: long EUR/USD at 1.0850, stop at 1.0825 — a 25-pip stop. A standard lot (100,000 units) makes each pip worth $10 in quote-currency terms, so 25 pips risks $250 per standard lot. $50 ÷ $250 = 0.2 lots. Take the trade at 0.2 lots and a full stop-out costs exactly the fifty dollars you chose — nothing more.

Worked example — crypto

Account: $5,000. Risk: 1% = $50. Trade: long BTC at 60,000 with the idea invalid below 58,800 — a $1,200 stop distance. $50 ÷ $1,200 = 0.0417 BTC (about $2,500 of exposure). The same calculation, run in reverse, is why leveraged “I’ll just size up” trades end accounts: the position grew, the stop distance did not, and the loss per trade quietly tripled.

Where traders cheat themselves

Widening the stop to keep the size. If the stop moves to fit a bigger position, the trade idea changed and the sizing maths is now theatre.
Risking “just this once” more. The 1% rule only compounds if it is actually a rule. The account does not care about your conviction.
Counting pips, not percent. Fifty pips means nothing without position size attached; the only universal unit is the percent of the account at risk.
Forgetting leverage is sizing, not capital. Leverage multiplies exposure, never the balance. The risk maths above is identical at 1:30 or 1:500 — only the margin required changes.

The part research can and cannot do

Position sizing cannot make a losing strategy win — it makes a losing strategy affordable to keep testing, which is how every strategy in our open research survived long enough to be judged. That is the honest scope of the tool: sizing keeps you in the game; the edge has to come from tested method. Our QUANTLAB framework and the VWAP mean-reversion lab exist because that second part deserves the same discipline as the first.