Built for Bettors Who Think in Edge
The Kelly Criterion is the mathematically optimal bet sizing formula for maximizing long-run bankroll growth. It was published by John Kelly Jr. at Bell Labs in 1956 and has been the backbone of professional gambling and quantitative investing ever since. Given two inputs — your true win probability and the decimal odds offered — Kelly tells you exactly what fraction of your bankroll to wager.
What Is the Kelly Criterion?
The Kelly formula is f* = (b·p − q) / b where b is the decimal odds minus 1 (the net payout per unit), p is your true win probability, and q is 1 − p. The result, f*, is the fraction of your bankroll to bet on this single wager. Bet too much and you blow up. Bet too little and you leave growth on the table. Kelly hits the optimum by maximizing the expected logarithm of your bankroll.
How to Use This Kelly Criterion Calculator
Enter your estimated win probability, the decimal odds, and (optionally) your bankroll. The calculator returns the Full Kelly stake, the dollar amount, your edge, and the expected per-bet bankroll growth. Use the no-vig calculator to derive a fair-probability estimate from sharp-book lines before plugging numbers in here.
Full Kelly vs Half Kelly vs Quarter Kelly
Full Kelly is the theoretically optimal stake under perfect probability estimates. In real-world betting, your probability is an estimate with error — and Full Kelly is unforgiving when you overestimate your edge. Half Kelly bets 50% of the recommended size, cutting variance dramatically while keeping ~75% of the growth rate. Quarter Kelly further halves the variance and is what most professional sports bettors actually use because it stays correct under realistic estimation error.
Kelly Criterion for Sports Betting
Kelly only works if you have a real, measurable edge. The hard part of profitable sports betting isn't the math — it's the probability estimate. Turtle +EV Labs surfaces the edge across 48 sportsbooks every 2 minutes, and Kelly tells you how much to bet on it. Pair this calculator with our arbitrage calculator for risk-free plays and our Pro picks platform for the model output that powers the probability input.
Common Mistakes When Using Kelly Criterion
- Treating implied odds as your true probability. If you plug in the book's implied probability, Kelly always says zero. You need an independent estimate.
- Using Full Kelly with a noisy model. Real edge estimates are noisy. Use Quarter or Half Kelly unless you have years of validated CLV data.
- Re-sizing only after wins. Kelly is a fraction of current bankroll — recompute the dollar stake after every settled bet, win or loss.
- Ignoring correlation. Two correlated bets at Full Kelly each can stack to over-bet your bankroll. Reduce stakes when bets aren't independent.
- Counting promo money as bankroll. Promo dollars often have rollover requirements that distort EV. Run Kelly only on your liquid, withdrawable bankroll.