Kelly Criterion

The Kelly criterion gives the size that grows wealth fastest in the long run for a given edge. The formula is small: the optimal fraction of capital to risk equals the edge divided by the odds — for a bet with win probability p and payoff b, Kelly is (p*b - (1-p)) / b. Plug in plausible trading numbers (say a 55 percent win rate at a 1:1 reward/risk) and full Kelly says risk 10 percent per trade.

Real traders almost never run full Kelly. The reason is variance. Full Kelly is optimal only when the edge and payoff are known exactly, which they never are in trading; even small estimation errors mean full Kelly bets are well above the actual growth-maximizing size and produce volatile equity curves the trader will not survive emotionally. The widespread practical rule is to run a fractional Kelly (typically a quarter to a half) on the trader's measured edge, which preserves most of the long-run growth while keeping drawdown distributions livable.

The deeper lesson Kelly bakes in is that sizing dominates entry over a career. The same setup, run with halved sizing, produces a meaningfully different equity curve because the variance scales with the square of the size and the drawdowns scale with the size itself. Position sizing is the lever; entries are the noise around it.