Not which trade to take — how much of your account to risk on it.
Position sizing is the decision of how much capital to put into any single trade, based on account size, risk tolerance, and how far away a stop-loss is set. It’s a different question from picking entries or reading trends — it governs how much you lose when a trade fails, not whether you were right to take it.
Most retail risk-management guidance caps a single trade at around 2% of total account value, precisely so that a string of losses — which will happen to everyone eventually — erodes the account gradually instead of wiping it out in one bad call.
The “2% rule” is a rounded-off, retail-friendly version of a real mathematical result. John Kelly’s 1956 Bell Labs paper on optimal bet sizing — the Kelly Criterion — showed there is a specific fraction of a bankroll that maximises long-run growth for a given edge and set of odds, and that betting more than that fraction doesn’t just add risk, it lowers your expected long-run return. It’s the modern, formalised version of the “gambler’s ruin” problem studied since the 17th century: past a certain bet size, going broke stops being a tail risk and becomes close to mathematically inevitable given enough trades.
Source: Kelly, J. L. (1956), “A New Interpretation of Information Rate,” Bell System Technical Journal.