Kelly Criterion Calculator

The Kelly criterion is a bankroll-sizing rule from 1956 information-theory work by John Kelly Jr. at Bell Labs. The formula answers one question — what fraction of a bankroll should be risked on a bet with a known edge to maximize the long-run growth rate of that bankroll? Bet too little and growth is slow. Bet too much and a bad run wipes everything out.

The formula itself is straightforward. The Kelly fraction equals the edge divided by the price, where edge is the bettor's estimated win probability multiplied by the net decimal odds, minus the loss probability. If a bet pays plus-100 American odds and the bettor estimates a 55 percent win rate, full Kelly recommends staking 10 percent of bankroll. That 10 percent figure assumes the 55 percent number is correct. If the true win rate is actually 52 percent, the same 10 percent stake is suddenly too aggressive and the bankroll's growth rate falls.

This sensitivity is why most professional bettors do not use full Kelly. They use a fraction of it — typically half or a quarter — for two reasons. The first is that win probability estimates are noisy. A bettor who thinks the edge is five percent might actually have a three percent edge, and full Kelly on the wrong number produces stake sizes that are too large. The second is that even when the estimate is correct, full Kelly produces drawdowns that are emotionally hard to ride out. A 50 percent peak-to-trough drop is normal at full Kelly in plausible scenarios. Half Kelly captures roughly three quarters of the growth rate while cutting drawdown depth substantially.

The calculator above returns full, half, and quarter Kelly stakes alongside the raw expected value of one dollar risked. When the recommended fraction is zero or negative, the bet is not positive expected value at the price offered — the market is paying less than the bettor's win probability suggests it should. That is not a sizing problem to fix with smaller stakes. That is a bet to skip entirely.