Skip to content
Omnikalc

Kelly criterion calculator: optimal bet size

Find the share of your bankroll to stake on a bet or investment with a positive edge, from the win probability and the odds, with fractional Kelly options.

Updated on
Your inputs
%

Value out of range

Odds format

Value out of range

$

Value out of range

Results

Stake as a share of bankroll

2.75%

Stake

$27.50

Full Kelly fraction
5.50%
Your edge (expected return per unit staked)
5.00%
Probability implied by the odds
52.38%
Expected growth of bankroll per bet
0.103%
Net odds (profit per 1 staked)
0.909

What the Kelly criterion answers

If you have an edge, a bet or investment where you expect to win more than you lose on average, how much should you stake? Bet too little and your bankroll grows slowly; bet too much and a run of losses can wipe you out even though every bet was favorable. In 1956, John Kelly of Bell Labs found the stake that maximizes the long-term growth rate of your bankroll.

The formula

f = p − q ÷ b*

  • p: your probability of winning
  • q = 1 − p: the probability of losing
  • b: net odds, the profit per $1 staked

If f* is zero or negative, the bet has no edge and Kelly says not to bet.

Converting odds

Format Example Net odds b
Decimal 2.10 2.10 − 1 = 1.10
American, negative −110 100 ÷ 110 = 0.909
American, positive +150 150 ÷ 100 = 1.5
Fractional 5/2 2.5

Worked example

You estimate a 55% chance of winning a bet offered at −110.

  • b = 0.909, q = 0.45
  • Full Kelly = 0.55 − 0.45 ÷ 0.909 = 5.5% of bankroll
  • The odds imply a 52.4% chance, so your edge is 5% per dollar staked
  • With half Kelly and a $1,000 bankroll, stake $27.50
  • Expected bankroll growth: about 0.10% per bet

Why most people bet a fraction of Kelly

Full Kelly is mathematically optimal only if your probability estimate is exactly right. In practice:

  • Estimates are noisy. Overestimating your edge by a few points can turn Kelly into overbetting, which lowers growth and raises the risk of ruin.
  • Swings are large. With full Kelly, there is a 50% chance that the bankroll will at some point fall to half its starting value.
  • Half Kelly keeps about three-quarters of the growth rate with roughly half the volatility; quarter Kelly is more conservative still.

Kelly in investing

The same idea applies to position sizing: with an expected excess return μ and variance σ², the Kelly fraction of capital in a risky asset is about μ ÷ σ². For stocks this often suggests leverage, which is why most investors use a small fraction of it, if any.

A word on gambling

Sportsbooks and casinos build a margin into their odds, so most bets have a negative expected value and Kelly says to stake nothing. Gambling can be addictive: if it stops being fun, call 1-800-GAMBLER for confidential help.

This calculator gives estimates for education and planning. It is not investment, tax or legal advice; for decisions about your own situation, talk to a qualified financial professional.

Frequently asked questions

What is the Kelly criterion?

A formula from John Kelly (1956) for the share of your bankroll to bet that maximizes long-term growth: f = p − q ÷ b, where p is the probability of winning, q = 1 − p and b the net odds.

Why do people use half Kelly?

Full Kelly is very volatile, and any overestimate of your edge leads to overbetting, which can be ruinous. Half Kelly keeps about three-quarters of the growth with far smaller swings.

What if the result is negative?

A negative Kelly fraction means the bet has no positive expected value at those odds, so the formula says not to bet at all.

How do American odds convert?

Positive odds such as +150 mean a $100 stake wins $150, so b = 1.5. Negative odds such as −110 mean you stake $110 to win $100, so b = 100 ÷ 110 ≈ 0.909.

Was this calculator helpful?

Thanks for your feedback!

Results are estimates provided for information purposes. They do not replace professional advice.