The Kelly Criterion Explained for Sports Bettors

You can find real edges and still go broke if you size your bets wrong. Kelly is the math that tells you how much to stake — and why the pros only ever use a fraction of it.

You can be a genuinely skilled bettor — finding real edges, beating the closing line — and still go broke. The missing piece is bet sizing: how much of your bankroll to put on each play. Bet too much and a normal losing streak wipes you out; bet too little and you barely grow. The Kelly criterion is the mathematical answer to sizing bets optimally. This guide explains what it is, how to use it, and why almost every pro uses a scaled-down version.

The problem Kelly solves

Imagine you have a real edge — a bet you believe wins 55% of the time at even money. It's clearly profitable. But how much should you stake? Bet your entire bankroll and you'll eventually hit the 45% and lose everything. Bet a single dollar and you're leaving nearly all your edge unrealized. Somewhere between "too much" and "too little" is an optimal amount that grows your bankroll as fast as mathematically possible without risking ruin. That optimal fraction is what the Kelly criterion calculates.

The Kelly formula

For a simple bet, the Kelly formula gives the fraction of your bankroll to wager:

f = (bp − q) ÷ b

Where f is the fraction of your bankroll to bet, b is the decimal odds minus 1 (your profit per unit staked), p is your estimated probability of winning, and q is the probability of losing (1 − p). If the result is zero or negative, Kelly is telling you there's no edge and you shouldn't bet at all.

That 10% might feel aggressive — and it is, which is the key insight we'll get to shortly. But first, notice what the formula captures: it scales your bet with both the size of your edge and the odds. Bigger edge, bigger bet. It also refuses to bet at all when you don't have an edge, which quietly protects you from your worst instincts.

Why full Kelly is too aggressive

Full Kelly maximizes growth if your probability estimate is exactly right. That's a big if. In sports betting you never know the true probability — you're estimating it, and your estimates will sometimes be off. When you overestimate your edge and bet full Kelly on it, you're overbetting, and overbetting is punished harshly by variance.

Full Kelly is also a wild ride even when your estimates are perfect. It produces enormous swings — drawdowns of 50% or more are routine. Most people can't stomach that emotionally, and emotional bettors make bad decisions. The combination of estimation error and brutal variance is why virtually no professional bets full Kelly.

The pro move: fractional Kelly

The standard solution is fractional Kelly — betting a fixed fraction of what the formula recommends, most commonly half Kelly or quarter Kelly. If full Kelly says bet 10%, half Kelly says bet 5%, and quarter Kelly says 2.5%.

The trade-off is remarkably favorable. Half Kelly captures about three-quarters of the growth rate of full Kelly while cutting the variance roughly in half. You give up a little growth for a large reduction in risk and a much smoother ride. Given that your probability estimates are imperfect anyway, that safety margin is exactly what you want. Quarter Kelly is even more conservative and is a sensible starting point for newer bettors or anyone unsure of their edge.

Where your probability comes from

Kelly is only as good as the win probability you feed it. Garbage in, garbage out — if you overestimate your edge, Kelly will faithfully tell you to overbet. This is where the rest of the toolkit matters. A disciplined way to estimate your edge is to compare the price you're getting against the fair, no-vig market price. If you can remove the vig from a sharp book's line and find that your bet's fair probability is higher than the price implies, that gap is your edge — and it's a far more grounded input than a gut feeling.

The Kelly calculator lets you enter your probability either as a percentage or as the fair odds you believe are correct, and it includes a guard that tells you when there's no edge to bet. It also offers full, half, and quarter Kelly side by side so you can see the trade-off directly.

Practical guidelines

• Define your bankroll and stick to it. Kelly is a fraction of your bankroll. Decide what that number is, and don't quietly redefine it after a loss.
• Use half or quarter Kelly. The reduced variance is worth the small growth cost, especially given estimation error.
• Be honest about your edge. The most common mistake is overestimating your probability. When in doubt, shade your estimate down.
• Recalculate as your bankroll changes. Kelly is proportional — as your bankroll grows or shrinks, your bet sizes should move with it.
• Don't bet without an edge. If the formula returns zero or negative, that's the answer. Pass.

The bottom line

Kelly turns bet sizing from a guess into a calculation. It scales your stake to your edge, protects you from ruin, and grows your bankroll as efficiently as the math allows. But its real-world use is fractional — half or quarter Kelly — because your edge is always an estimate and full Kelly punishes estimation error severely. Get your probability from a disciplined process like devigging the market, size with fractional Kelly, and you've got the two halves of a professional staking approach: knowing your edge, and knowing what to do with it.