Most sports bettors focus entirely on which team to back. Far fewer think carefully about how much to stake, and that's where the majority of bankroll damage happens. Bet too little and you leave value on the table. Bet too much and a bad run wipes you out before the edge pays off.
The Kelly Criterion solves this problem. It's a mathematical formula that calculates the precise fraction of your bankroll you should bet on any wager, based on your estimated edge over the odds. This guide explains what it is, how the formula works, how to apply it to sports betting on Duelbits, and why most professionals use a modified version.
In probability theory, the Kelly Criterion is a formula for risk allocation, sizing a sequence of bets by maximising the long-term expected value of the logarithm of wealth, which is equivalent to maximising the long-term expected geometric growth rate. John Larry Kelly Jr., a researcher at Bell Labs, described the criterion in 1956.
The formula was originally purposed to help Kelly's company reduce long-distance telephone signal noise issues. In time, however, sports bettors and Wall Street traders realised the value of the formula as an investment management system that could help protect assets and maximise earnings. The Kelly Criterion is now widely recognised as the most mathematically optimal sports betting bankroll management strategy.
In practical terms: Kelly tells you how much of your bankroll to bet when you believe you have an edge. The bigger the edge, the bigger the recommended stake. When there is no edge, Kelly recommends betting nothing.
The Kelly Criterion is a mathematical formula for determining the optimal stake amount that provides sports bettors with a way of maximising the growth rate of their bankroll over time while minimising losses. Instead of treating every dollar in your bankroll equally, the Kelly Criterion values growth in proportion to your current bankroll.
The Kelly formula is:
f = (bp − q) / b
Where:
f = the fraction of your bankroll to bet b = decimal odds minus 1 (your net profit on a 1-unit win) p = your estimated probability of winning (expressed as a decimal) q = your estimated probability of losing (= 1 − p)
For example, with a 55% win probability and decimal odds of 2.10 (b = 1.10): f = (1.10 × 0.55 − 0.45) / 1.10 = 0.0136 or 1.36% of bankroll.
The situation: You're betting on a football match. The sportsbook has priced Team A at 2.10 decimal odds. After your own analysis, you believe Team A has a 57% chance of winning, higher than the 47.6% implied by the odds. You've identified a positive expected value bet.
Inputs:
Calculation: f = (1.10 × 0.57 − 0.43) / 1.10 f = (0.627 − 0.43) / 1.10 f = 0.197 / 1.10 f = 0.179 = 17.9%
Kelly recommends staking 17.9% of your bankroll on this bet.
If your bankroll is $1,000, that's a $179 stake. If your edge estimate is correct and you make many similar bets over time, Kelly's formula maximises your long-term bankroll growth rate.
If the Kelly percentage results in less than 0%, the Kelly Criterion recommends walking away and not betting at all, the odds do not seem to be in your favour based on the formula and mathematical calculation.
A negative result means your estimated win probability is lower than the probability implied by the odds. The bet has negative expected value. Kelly is not just a staking system, it also functions as a filter, telling you when a bet is worth placing at all.
This is one of the Kelly Criterion's most valuable properties for sports bettors: it stops you from betting on games where you don't have a genuine edge.
Full Kelly betting, staking exactly what the formula recommends, is mathematically optimal, but it produces significant bankroll swings in practice. A Kelly bettor has a 1/3 chance of halving their bankroll before doubling it. A half-Kelly bettor only has a 1/9 chance of halving their bankroll before doubling it.
Even Kelly supporters usually argue for fractional Kelly, betting a fixed fraction of the amount recommended by Kelly, for a variety of practical reasons, such as wishing to reduce volatility, or protecting against non-deterministic errors in their edge calculations.
The most common approaches:
In the example above, half Kelly would recommend 8.95% ($89.50 on a $1,000 bankroll) instead of the full 17.9%.
| Staking Method | Stake Size | Adapts to Edge? | Adapts to Bankroll? | Ruin Risk |
|---|---|---|---|---|
| Flat betting (fixed %) | Fixed | No | No | Low |
| Fixed units | Fixed | No | No | Low-Medium |
| Martingale | Doubles after loss | No | No | Very High |
| Kelly Criterion | Variable | Yes | Yes | Low (if correct) |
| Fractional Kelly | Variable (reduced) | Yes | Yes | Very Low |
The Kelly Criterion is considered one of the most effective money management strategies for many types of gamblers, including sports bettors. It offers an excellent way to avoid the common trap of betting too much and blowing your bankroll.
Unlike flat betting, Kelly automatically scales your stakes up as your bankroll grows and down when it shrinks. This compounding mechanic is what makes it mathematically superior over a long betting career.
To apply the Kelly Criterion strategy, bettors must accurately assess the probability of their bet winning. This requires thorough analysis and a deep understanding of the sport and factors affecting the outcome. It's crucial to note that overestimating your edge can lead to overbetting and potential bankroll depletion.
The Kelly formula is only as good as the probability estimate you feed into it. If you estimate 57% and the true probability is 52%, Kelly will recommend a stake size that is too large, and over many bets, this leads to worse outcomes than more conservative staking.
Most professionals use fractional Kelly (25% or less) to account for estimation errors and reduce variance. Think of the fractional multiplier as a built-in safety margin for the inevitable imprecision in your probability estimates.
Kelly is most appropriate for bettors who:
1. Overestimating your edge. The single most damaging error. If you believe you have 60% probability when the true figure is 52%, Kelly will systematically oversize your stakes and erode your bankroll.
2. Using full Kelly without understanding the volatility. Full Kelly swings are severe. A losing run can feel catastrophic even when the long-term expectation is positive. Most bettors cannot handle this emotionally.
3. Applying Kelly to every bet without discrimination. Kelly is a filter as much as a staking formula. If it returns zero or negative, that's the formula telling you not to bet, not a signal to switch to flat betting as a workaround.
4. Ignoring the effect of the vig. Sportsbook margin means that implied probabilities across a market add up to more than 100%. Always calculate Kelly based on the true probability you estimate, not the implied probability from the odds.
Here's how to apply Kelly to a real bet on Duelbits Sportsbook:
You see a tennis match priced at 2.25 (implied probability 44.4%) for the underdog. Your analysis, based on surface stats, head-to-head record, and fitness information, gives the underdog a 50% chance. That's a genuine edge.
Kelly calculation:
f = (1.25 × 0.50 − 0.50) / 1.25 f = (0.625 − 0.50) / 1.25 f = 0.125 / 1.25 f = 0.10 = 10%
At half Kelly: 5% of your bankroll.
On a $500 bankroll, that's a $25 stake, a disciplined, calibrated bet that reflects both your confidence level and the size of your edge.
What is the Kelly Criterion? The Kelly Criterion is a mathematical formula that calculates the optimal fraction of your bankroll to stake on a bet, based on your estimated edge over the odds. Developed in 1956, it maximises long-term bankroll growth while minimising ruin risk.
What is the Kelly Criterion formula? f = (bp − q) / b. Where f = fraction of bankroll to bet; b = decimal odds minus 1; p = estimated win probability; q = estimated loss probability (1 − p).
What is fractional Kelly betting? Staking a fixed fraction (typically 50% or 25%) of the full Kelly recommendation, to reduce bankroll volatility and protect against probability estimation errors. The most commonly used approach by professional bettors.
What does a negative Kelly result mean? No edge, the bet has negative expected value. Kelly recommends not betting at all when the result is zero or negative.
What is the biggest risk of using the Kelly Criterion? Overestimating your edge. If your probability estimate is too high, Kelly recommends stakes that are too large, leading to overbetting and bankroll damage over time.
Is the Kelly Criterion the same as flat betting? No. Flat betting uses a fixed stake on every bet regardless of edge size. Kelly varies the stake based on the size of the edge and automatically scales with your bankroll.
Who invented the Kelly Criterion? John L. Kelly Jr., a researcher at Bell Labs, in 1956. Originally developed for telephone signal processing, it was later adopted by gamblers and investors. It is sometimes called "Fortune's Formula."
Can I use the Kelly Criterion on every bet? You can apply it to any bet, but it requires an honest probability estimate. Without a genuine edge, Kelly returns zero or negative and recommends not betting. It works best for experienced bettors who systematically identify value.
