Enter your bankroll, estimated win probability, and the American odds. The calculator returns the bankroll-optimal bet size at full, half, and quarter Kelly — with a sanity cap for survival.
The Kelly Criterion answers one question: given a bet with a known edge and a known payoff, what fraction of your bankroll maximizes long-run growth? The formula is f* = (bp − q) / b, where b is decimal odds minus 1, p is your win probability, and q is 1 − p.
The output (% of bankroll) tells you what fraction Kelly says to bet. The bet-size dollar figure applies that fraction to the bankroll you entered. The "edge vs market" line shows how many implied-probability points your win-probability estimate beats the price by — that's the entire reason Kelly is positive in the first place. If your edge is negative, Kelly outputs zero (the math says don't bet).
Full Kelly is mathematically optimal only if your probability estimate is exact. In real-world sports betting, every estimate has noise. Full Kelly massively overbets when your estimates are off, which produces drawdowns that destroy bankrolls before the long-run math can save you.
Half Kelly produces roughly 75% of full Kelly's growth with materially less drawdown risk. Quarter Kelly is more conservative still and the right default for noisy estimates. For a deeper walk-through of why this matters: Kelly Criterion for real bankrolls: why fractional Kelly beats flat unit betting →