The Kelly Criterion is the bet-sizing formula that maximizes the expected logarithmic growth rate of your bankroll. It tells you exactly how much of your bankroll to risk on each +EV bet so you grow as fast as possible without going broke.
Most professional sports bettors use Half Kelly or Quarter Kelly, not Full Kelly. The reason: your true edge is almost always overestimated, and Full Kelly is brutally volatile. Half Kelly retains roughly 75% of the long-term growth rate at a fraction of the variance.
Formula
Kelly fractionf = (bp β q) / b
Whereb = decimal odds β 1, p = win prob, q = 1 β p
Frequently asked questions
Should I bet Full Kelly?
Almost never. Full Kelly assumes you know your edge precisely. Real-world edges are noisy estimates, so Full Kelly tends to over-bet and produce huge drawdowns. Half Kelly is the standard sharp-bettor default; Quarter Kelly is appropriate when your edge estimate is uncertain.
What if Kelly tells me to bet 0% or negative?
That means the wager is -EV at the odds you entered β Kelly says don't bet at all. The calculator will show no recommendation when there's no edge.
Can I use Kelly with parlays?
Technically yes, but the variance is severe and your edge is hard to estimate. Most bettors apply Kelly to straight bets only and stake parlays at fixed flat amounts.
Why does the calculator show growth rate, not just bet size?
The whole point of Kelly is maximizing logarithmic growth. Showing the growth rate at each fraction lets you see why Half Kelly is the better risk-adjusted choice β you give up almost nothing in expected growth but cut variance dramatically.
What's the relationship between Kelly and EV?
EV measures expected dollar profit per bet. Kelly maximizes long-run bankroll growth given that EV. Two bets can have identical EV but very different optimal Kelly stakes depending on the win probability and odds.