Kelly Criterion Calculator

About this tool

Two people trade the same strategy: one sizes too big and gets knocked out by a drawdown, the other sizes too small and barely grows. The Kelly formula turns win rate and payoff ratio into a single fraction that answers how much to risk for the fastest long-run growth, not whether the next trade wins.

At a 55% win rate, a 1.5 payoff ratio and a 100k bankroll, full Kelly is 25%, half Kelly is 12.5%, the stake is 12,500, the edge is 0.275 and geometric growth is about 1.85% per round. Drop the win rate to 45% and Kelly turns negative, so no stake is recommended.

What it does

• Win rateWin rate entered as a percentage Every number shown here maps to a field on the form, so changing one value recomputes the rest. The units are printed next to each input, which saves most of the usual mistakes.
• PayoffPayoff ratio from average win and loss The main figures sit at the top of the result panel and the breakdown sits below it, so you can read the conclusion first and audit the steps afterwards. Nothing you type leaves the browser.
• FractionThree fractional Kelly settings All arithmetic runs locally in the page. There is no request to a server, no account, and no saved history once the tab is closed. Use the example button if you want a set of values that is known to work.
• NegativeNo stake when Kelly is negative Intermediate quantities are kept in the detail table. When a total looks wrong, tracing back through those rows is much quicker than staring at the headline number alone.
• GrowthLong-run geometric growth shown The formula is written out in the notes rather than hidden behind a label, so you can check it against a spreadsheet or a textbook. Values are rounded only for display.
• BreakdownBreakdown lists p, q and b Inputs are validated before any calculation starts. Empty or impossible values produce a short explanation instead of a zero, a NaN, or a made-up answer.

How to use it

1. Start by filling inenter your historical win rate The main figures sit at the top of the result panel and the breakdown sits below it, so you can read the conclusion first and audit the steps afterwards. Nothing you type leaves the browser.
2. Next, setdivide average win by average loss All arithmetic runs locally in the page. There is no request to a server, no account, and no saved history once the tab is closed. Use the example button if you want a set of values that is known to work.
3. Then pressenter the current bankroll Intermediate quantities are kept in the detail table. When a total looks wrong, tracing back through those rows is much quicker than staring at the headline number alone.
4. Read thepick full or fractional Kelly The formula is written out in the notes rather than hidden behind a label, so you can check it against a spreadsheet or a textbook. Values are rounded only for display.
5. Finally, usepress Calculate and read the stake Inputs are validated before any calculation starts. Empty or impossible values produce a short explanation instead of a zero, a NaN, or a made-up answer.

Things to keep in mind

• Rangewin rate must be between 0 and 100 All arithmetic runs locally in the page. There is no request to a server, no account, and no saved history once the tab is closed. Use the example button if you want a set of values that is known to work.
• Positivepayoff ratio must be positive Intermediate quantities are kept in the detail table. When a total looks wrong, tracing back through those rows is much quicker than staring at the headline number alone.
• Samplea small sample distorts the win rate The formula is written out in the notes rather than hidden behind a label, so you can check it against a spreadsheet or a textbook. Values are rounded only for display.
• Single roundthe result is a long-run figure Inputs are validated before any calculation starts. Empty or impossible values produce a short explanation instead of a zero, a NaN, or a made-up answer.
• Live tradingtraders usually stay at half Kelly or below Every number shown here maps to a field on the form, so changing one value recomputes the rest. The units are printed next to each input, which saves most of the usual mistakes.