Kelly criterion calculator (and why half-Kelly)
There is one bet size that grows an account fastest. Bet more and you grow slower, not faster. Put in your real numbers and see where that line is.
| Bet size | Risk per trade | Growth vs full Kelly | After 5 losses in a row |
|---|
How it works: the payoff is your average win ÷ your average loss. Kelly % = win rate − (loss rate ÷ payoff). It is the share of the account to risk on each trade that grows it fastest, if your numbers are exact and never change. If that comes out at 0 or less, you have no edge, and Kelly says risk nothing. "Growth vs full Kelly" compares how fast each bet size grows the account per trade on average, with full Kelly as 100%. The last column is what 5 losses in a row would cost at that size.
Why almost nobody bets full Kelly
Full Kelly is the fastest-growing bet only if your win rate and sizes are exactly right. Yours are a guess from past trades, and a small sample can be way off. Bet more than Kelly and growth slows down. At about twice Kelly it stops. Half Kelly gives up only about a quarter of the growth and cuts every swing in half. That trade is worth it. Losing streaks hurt less, and a wrong guess about your edge does far less damage. See the Kelly criterion in plain words.
Kelly is a ceiling, not a target
Most traders who use Kelly treat it as the most they'd ever risk, not what they risk every time. If Kelly says 15%, that is a sign your numbers look strong, not a reason to bet 15%. Small, fixed risk is easier to live with and easier to stick to. Start with how much to risk per trade, check your numbers with the expectancy calculator, and see what a bet size does to you over time with the risk of ruin calculator.
Make this a habit, not a one-off. Lesson 2 of our free email course is all risk: the 1% rule, where your stop goes, and the 2-loss rule. Send me the free lessons →
Follow along