Kelly Criterion Calculator

Optimal bet sizing for your bankroll. Enter the odds, your true probability of winning, and your bankroll — get the growth-maximizing bet size, plus half- and quarter-Kelly recommendations for variance-conscious sizing.

Odds format
Implied: 40.00%
%

Between 0 and 100.

$
Optimal sizing
Your edge
+5.00%
Full Kelly: 8.33% of bankroll
Full Kelly
$83.33
growth-max, high variance
Half Kelly
$41.67
balanced — recommended
¼ Kelly
$20.83
conservative
Read the guide

Bankroll Management: Why Most Bettors Blow Up

Why bad sizing kills more bankrolls than bad picks. Unit math, Kelly basics, the psychology of variance, and a survival framework.

Read it
Made To Parlay

We think in parlays. Multi-leg bets where every leg has to hit. The same logic powers our app: every day is a parlay of habits, and the streak only counts if every leg cashes.

See the app

Kelly criterion FAQ

What is the Kelly criterion?

A formula that tells you the bankroll fraction to bet on any given +EV opportunity to maximize long-run growth. Developed by John Kelly Jr at Bell Labs in 1956. Used by professional bettors, hedge fund managers, and anyone who needs to size repeated bets with edge.

How does the math work?

f* = (b × p − q) / b, where b is the decimal odds minus 1 (the multiple of your stake you win on success), p is your true probability of winning, and q = 1 − p. The result is the fraction of your bankroll to risk on this bet.

What's the difference between full, half, and quarter Kelly?

Full Kelly is mathematically growth-optimal but extremely high variance — drawdowns of 50%+ are routine. Half Kelly cuts variance dramatically (about half the volatility) at the cost of slightly slower growth. Quarter Kelly trades even more growth for stability — most pros use somewhere between half and quarter Kelly in practice.

What if my Kelly is negative?

Negative Kelly means the bet is -EV — your probability is lower than the implied probability priced by the book. The calculator shows 'Don't bet' in that case. Kelly is silent on which way to bet; you'd have to take the *opposite* side at the matching odds (if available) to flip a -EV play into +EV.

Do I really know my probability?

This is the catch. Kelly assumes you have a true probability estimate that beats the book. Most bettors overestimate their edge — meaning they over-bet via Kelly. The conservative move is to trust half or quarter Kelly even if you think your estimate is exactly right.

Can I use Kelly across multiple bets at once?

Multi-bet Kelly is more complex than the single-bet formula. The simple version: independent bets can be sized independently, but bets that share correlated outcomes (same game, same market) need joint optimization. For most retail bettors, treat each bet independently with fractional Kelly and you'll be fine.