For bettors

Kelly criterion calculator

The stake size that maximises long-run bankroll growth for a given edge — with fractional-Kelly options for a smoother ride.

Inputs

Decimal odds

Your estimated win probability52%

The market implies 47.6%. Kelly only bets when your estimate is higher.

Bankroll

Kelly fraction

Full Kelly is aggressive and swingy. Most bettors use half or quarter.

Result

Stake

—

% of bankroll

—

Full-Kelly fraction—

Your edge—

Expected value per unit staked—

Kelly assumes you know your true edge — you almost never do. Overestimating probability makes Kelly bet far too big. Fractional Kelly and a hard per-bet cap protect against that. This is not betting advice.

What the Kelly criterion does

The Kelly criterion answers one question: given an edge, what fraction of your bankroll should you stake to grow it fastest over the long run without going broke. The formula is f* = (b·p − q) ÷ b, where b is the decimal odds minus 1, p is your estimated win probability and q is 1 − p. Bet more than f* and your long-run growth rate actually falls while volatility climbs; bet less and you grow more slowly but far more smoothly.

Why almost nobody uses full Kelly

Full Kelly assumes your probability is exactly right. It never is. A 5-point overestimate can turn the "optimal" stake into a losing strategy, and even with a correct edge full Kelly routinely draws the bankroll down 50% or more before recovering. Half Kelly keeps about three-quarters of the growth rate for roughly half the volatility; quarter Kelly is the common choice for anyone whose probabilities are estimates rather than known.

When Kelly says stake nothing

If f* comes out at zero or below, your probability is not higher than the price's implied probability, so there is no edge and any stake is negative expected value. That is the tool working correctly, not a bug — Kelly will not size a bet it cannot justify.

Frequently asked questions

What is the Kelly criterion formula?

f* = (b·p − q) ÷ b, where b is decimal odds minus 1, p is your win probability and q is 1 − p. If f* is zero or negative, Kelly says do not bet.

Should I use full Kelly?

Rarely. Full Kelly is highly volatile and unforgiving of probability errors. Half or quarter Kelly keeps most of the growth with far smaller swings.

Why does Kelly recommend no bet?

Because your estimated win probability is not higher than the price's implied probability — there is no edge, so any stake is negative expected value.

How much does a wrong probability estimate hurt?

A lot. Kelly is quadratically sensitive near the optimum: overestimating your edge by a third can flip a growth strategy into a slow decline. Betting a fixed fraction of full Kelly (half or quarter) is the standard defence against this.

Does Kelly apply to accumulators and multiple simultaneous bets?

The simple formula assumes one bet at a time with a binary outcome. For several bets running at once — especially if they are correlated — you need the multivariate version; using single-bet Kelly on each will over-stake your bankroll in total.

Should the bankroll I enter be my whole net worth?

No. Use the amount you have genuinely set aside for betting and can treat as a single pool. Kelly fractions of your life savings or money you need for something else are not what the model assumes.

Why is the recommended stake sometimes tiny?

Because your edge is small. A 1-point edge at even money is a Kelly fraction of about 1%. Small edges are normal; the model is telling you that betting big into a thin edge is how bankrolls die.

Does the calculator store my bankroll figure?

No. Everything is computed in your browser and nothing is transmitted or saved. Reloading the page clears it.

All tools →