Definition
What is the Kelly criterion?
Last updated
Definition
The Kelly criterion is a formula for sizing bets to maximize the long-run growth rate of a bankroll. For a bet with probability *p* of winning at decimal odds *o*, full-Kelly size is (p × o − 1) / (o − 1) as a fraction of bankroll.
Full Kelly is mathematically optimal but operationally suicidal: variance with full-Kelly sizing is enormous and one mis-estimated edge can blow the bankroll. Real operators use fractional Kelly — typically 0.25× to 0.5× — which trades some long-run growth for survivable variance. Glitch Edge defaults to 0.25× Kelly with a max-per-bet bankroll cap.
The formula
Kelly fraction f* = (p × o − 1) / (o − 1)
Where p = probability of winning (your model’s estimate), o = decimal odds being offered. The result is the fraction of bankroll to stake on a single bet.
Why full Kelly is rare
Full Kelly maximizes long-run log-growth but the path is brutal — drawdowns of 50%+ are common even on a real edge. Most operators run 0.25× Kelly (variance per bet ≈ 1/16 of full Kelly) and accept the slower growth in exchange for surviving the variance.
In Glitch Edge
The platform refuses Kelly fractions above 1× regardless of operator config. Default is 0.25× with a 2% max-per-bet bankroll cap.