Kelly Criterion

The Kelly criterion is a formula for the bet size that maximises long-run growth of capital given the probability of winning and the payoff ratio; in practice investors use a fraction of it. It is closely related to Position Sizing, Expected Value, Calibrated Confidence and Maximum Drawdown, each defined in the hedgewing.ai glossary.

Developed by John Kelly at Bell Labs in 1956, the Kelly criterion computes the fraction of capital to risk on a favourable bet so that the expected logarithm of wealth, and therefore the long-run growth rate, is maximised. For a simple bet with win probability p and net odds b (win b for every 1 risked), the optimal fraction is f = p - (1 - p) / b. With a 55% win probability and even odds, Kelly says risk 10% of capital.

Full Kelly is theoretically optimal but practically harsh: it assumes the probability and payoff are known exactly, and it produces drawdowns that most people cannot tolerate. Because real estimates are noisy and overestimating the edge leads to overbetting, which lowers growth and can be ruinous, practitioners commonly use half-Kelly or quarter-Kelly. These fractions give up a small amount of theoretical growth for a large reduction in variance.

The Kelly framework is a useful way to think about calibrated forecasts: a tool that reports an honest probability lets you compute a Kelly fraction, while a tool that reports inflated confidence will cause systematic overbetting. hedgewing.ai's calibrated confidence is intended to be a usable input to this kind of sizing, with the strong caveat that fractional Kelly, a hard per-position risk cap, and attention to correlation across positions should always override the raw formula.

FAQ

What is Kelly Criterion?

The Kelly criterion is a formula for the bet size that maximises long-run growth of capital given the probability of winning and the payoff ratio; in practice investors use a fraction of it.

Which terms are related to Kelly Criterion?

Kelly Criterion is closely related to Position Sizing, Expected Value, Calibrated Confidence and Maximum Drawdown. Each has its own entry in the hedgewing.ai glossary.

Is this definition of Kelly Criterion investment advice?

No. This glossary entry explains what Kelly Criterion means. It is educational reference material, not personalised investment advice and not a recommendation to buy or sell any security.

Create a free Hedgewing account to try hedgewing.ai yourself.