Kelly Criterion Calculator

Calculate the optimal bet or position size given your win probability and payout.

How this is calculated

The Kelly formula is f* = (p × b − q) / b, where p is your win probability, q is 1 − p, and b is your win multiple. When win and loss multiples differ, the generalised form f* = (p × win − q × loss) / (win × loss) is used instead. The result is the fraction of your bankroll that maximizes long-run geometric growth.

If the calculated fraction comes out negative, the bet has negative expected value and Kelly recommends staking nothing. Because real-world edge estimates are rarely perfectly accurate, most practitioners bet a fraction of full Kelly — commonly half — to reduce volatility and protect against overestimating their edge.

Frequently asked questions

What is the Kelly Criterion?
A formula that calculates the optimal fraction of your bankroll to bet or invest in a favorable opportunity, in order to maximize the long-run geometric growth rate of your capital. Developed by John Kelly at Bell Labs in 1956, it is used in gambling, sports betting, and position sizing for trading.
Why use half-Kelly instead of full Kelly?
Full Kelly maximizes long-run growth but comes with large swings in bankroll value, and is very sensitive to overestimating your true edge. Half-Kelly gives up some growth rate in exchange for significantly less volatility and more protection against estimation error — a trade-off most practitioners consider well worth it.
What is the "win multiple"?
How many dollars you receive per dollar wagered if you win, not counting your original stake. A "2.0" win multiple (2:1 odds) means a $100 bet that wins pays out $200 in profit, on top of getting your $100 stake back.
Can Kelly be used for stock trading?
Yes, though estimating a reliable win probability and payout ratio for a trading strategy is much harder than for a game with known odds. Many traders use a fraction of Kelly (often much less than half) specifically because their edge estimates carry meaningful uncertainty.
What happens if I bet more than the Kelly fraction?
Betting above full Kelly ("over-betting") actually reduces your long-run growth rate, not just your risk — beyond a certain point, more aggressive sizing makes both the ride rougher and the destination worse. This is one of Kelly's more counter-intuitive results.
Is Kelly appropriate for a full investment portfolio?
The classic formula assumes a single repeated bet with known, fixed odds — real portfolios involve many correlated positions with uncertain and changing probabilities, so Kelly is better used as a sizing principle (bet less when your edge or confidence is smaller) than a literal formula for portfolio construction.
What does negative expected value mean?
If your expected value per bet is negative, there is no bet size that makes the wager profitable in the long run — the mathematically optimal Kelly fraction is 0%, meaning you should not take the bet at all, regardless of how good the odds look on any single try.
Can I use this in my own app?
Yes — every calculator on Stupidly Clever has a matching REST API and MCP tool that runs the same underlying logic.

Related calculators