Kelly Criterion for Sports Betting: The Complete Guide

Most bettors spend all their time deciding what to bet on and almost none deciding how much to bet. That’s a mistake. Even a bettor with a genuine, provable edge can go broke if they size their stakes badly, and a mediocre bettor can survive far longer than they should by staking too conservatively. The Kelly Criterion is the mathematical answer to that second question: given your edge, exactly how much of your bankroll should you risk?

This guide breaks down what the Kelly Criterion is, how the formula works, and how to apply it practically using our free Kelly Criterion Calculator.


What Is the Kelly Criterion?

The Kelly Criterion is a formula developed in 1956 by John L. Kelly Jr., a scientist at Bell Labs, originally to solve a problem in signal transmission. Gamblers and investors later realized it solved a much more useful problem: how to size bets so that a bankroll grows as fast as mathematically possible, while avoiding the risk of ruin that comes from overbetting.

Unlike flat staking (betting the same amount every time) or arbitrary “gut feel” sizing, Kelly ties your stake directly to two things: how big your edge actually is, and what odds you’re being offered. A bigger edge means a bigger recommended stake. A smaller edge, or no edge at all, means a smaller stake, or no bet.


The Kelly Formula Explained

The classic Kelly formula looks like this:

f* = (bp – q) / b

  • f is the fraction of your bankroll to stake
  • b is the net odds received on the bet (decimal odds minus 1)
  • p is your estimated probability of winning
  • q is the probability of losing (1 – p)

It looks intimidating written out, but the logic behind it is simple: bet more when the odds and your win probability both favor you, and bet nothing when they don’t.


A Worked Example

Say a bookmaker offers decimal odds of 2.50 on a team you believe has a 45% chance of winning: higher than the odds imply (an implied probability of 40%). That gap between your estimate and the market’s is your edge.

Plugging in the numbers: b = 1.50, p = 0.45, q = 0.55.

f = (1.50 × 0.45 – 0.55) / 1.50 ≈ 0.083, or about 8.3% of your bankroll.

That’s the full Kelly stake for this specific bet. Rather than doing this arithmetic by hand every time, our Kelly Criterion Calculator does it instantly: just enter the odds and your estimated win probability.


Full Kelly vs. Fractional Kelly

Here’s the catch: 8.3% on a single bet is aggressive. Full Kelly assumes your probability estimate is exactly correct, and in real sports betting, it almost never is: your 45% might really be 42% or 48%. Because Kelly is highly sensitive to overestimated edges, most professional bettors don’t use full Kelly at all. Instead, they use a fraction of it.

Half Kelly (staking 50% of the full Kelly amount) is the most common approach. It sacrifices some theoretical long-term growth in exchange for a dramatically smoother bankroll curve and much less damage from a bad probability estimate. In the example above, that would mean staking about 4.2% instead of 8.3%.

Some bettors go even more conservative and use quarter Kelly, especially when their model or data source is newer and less proven. There’s no universally “correct” fraction. It depends on how confident you genuinely are in your probability estimates.


Kelly Criterion vs. Flat Betting

Flat betting (staking the same fixed amount or the same fixed percentage on every bet regardless of edge) is easier to manage, but it treats a small edge and a large edge identically. A bettor using flat stakes leaves growth on the table when their edge is large, and stays overexposed when their edge is thin or nonexistent.

Kelly staking, by contrast, scales the bet to the opportunity. That’s precisely why it’s the standard reference point in professional bankroll management, even among bettors who ultimately choose to flatten it out with a fractional approach rather than follow it exactly.

Simulation: What Happens If You Overestimate Your Edge

Kelly is only as good as your probability estimate. We simulated 20,000 bettors per row who bet at odds of 2.0 on outcomes with a true win chance of 52%. Some estimate correctly, some are 3 or 6 percentage points too optimistic: a very common mistake. Each places 1,000 bets:

Your estimateStakingStake per betTypical resultEnds below startBankroll halves along the way
52% (true 52%)Full Kelly4.0%×2.2327.6%38.3%
52% (true 52%)Half Kelly2.0%×1.8217.8%7.3%
52% (true 52%)Quarter Kelly1.0%×1.4213.4%0.1%
55% (true 52%)Full Kelly10.0%×0.3663.6%87.2%
55% (true 52%)Half Kelly5.0%×2.1232.2%51.6%
55% (true 52%)Quarter Kelly2.5%×1.9919.7%14.6%
58% (true 52%)Full Kelly16.0%almost 090.4%98.3%
58% (true 52%)Half Kelly8.0%×1.0051.9%77.5%
58% (true 52%)Quarter Kelly4.0%×2.2327.9%37.8%
Monte Carlo simulation, 20,000 runs per row, 1,000 bets each. Typical result = median final bankroll as a multiple of the start.

With a correct estimate, full Kelly grows fastest, but it halves the bankroll along the way in more than a third of cases. Overestimate your chance by just 3 points, and full Kelly turns a winning edge into a typical loss of about two thirds. Quarter Kelly with a 6-point error still ends with the same typical result as full Kelly with a perfect estimate. That is why most professionals use half or quarter Kelly: our Kelly Calculator shows both.



Common Kelly Criterion Mistakes

  • Overestimating your win probability. Kelly is only as good as the number you feed it. An inflated edge produces an oversized, dangerous stake.
  • Using full Kelly on every bet. Without a fractional adjustment, normal estimation error can lead to large bankroll swings.
  • Ignoring correlated bets. Staking full Kelly-recommended amounts on multiple correlated outcomes (like several bets on the same game) compounds risk in ways the basic formula doesn’t account for.
  • Recalculating from a shrinking bankroll without adjusting. Kelly stakes are a percentage of your current bankroll, not a fixed dollar amount. They should shrink and grow as your bankroll does.

Why Bet Sizing Matters as Much as Bet Selection

Finding value is only half the job. Understanding Expected Value (EV) in Sports Betting tells you which bets are worth making: the Kelly Criterion tells you how much to put on them once you’ve found one. Skip the sizing step, and even a genuinely profitable long-term strategy can wipe out your bankroll through a string of oversized bets during a losing streak. That’s why bet sizing sits right alongside the fundamentals covered in our Bankroll Management guide as one of the most important, and most overlooked, skills in sports betting.


Try the Kelly Criterion Calculator

Rather than running the formula by hand for every bet, use our free Kelly Criterion Calculator. Enter the odds and your estimated win probability, and it instantly returns the recommended stake, including half and quarter Kelly options for a more conservative approach.

Used consistently and honestly, with realistic probability estimates rather than wishful ones, the Kelly Criterion is one of the simplest ways to turn a genuine edge into long-term bankroll growth, without the wild swings that come from guessing your stake size.