How the Kelly criterion works
The Kelly criterion is a formula for the share of a bankroll that maximizes long-run growth when you have an edge. It needs your own estimate of the win probability, and that is the weak point: a small error in the estimate changes the answer a lot.
With no edge (your probability is at or below the price’s break-even rate), Kelly is zero: it never says to bet the other side. Many people stake only a fraction of Kelly, such as half or a quarter, to cut the swings.
- Full Kelly fraction = p − (1 − p) ÷ b, where p is your win probability and b is the decimal odds minus 1.
- Fractional Kelly = the full fraction × a multiplier (0.5 for half Kelly, 0.25 for a quarter).
- Stake = fraction × bankroll.
Questions
What is the Kelly criterion?
A formula from probability theory that gives the fraction of a bankroll to stake on a bet with a positive expected value so as to maximize long-run growth.
Is full Kelly a good idea?
Full Kelly assumes your probability estimate is exactly right and accepts large swings. Because estimates are uncertain, many bettors use half or quarter Kelly. Nothing here is advice to bet.
What does it mean when Kelly is 0?
Your estimated win probability does not beat the break-even rate for the price, so the formula sees no edge and recommends no stake.
Where do I get a win probability?
From your own research or model. It is not provided by the price: the price’s implied probability is only the break-even rate. Hit rates need a large, current sample to be useful.
Is this gambling advice?
No. It is an educational calculator. Betting involves risk, you can lose all you wager, and Livdy is research, not a sportsbook.
Put it into practice: Player prop research. Livdy shows every book’s price side by side and the game logs behind every prop hit rate.