Kelly Criterion: The Complete Guide to Optimal Bet Sizing

The Original Kelly Formula

John L. Kelly Jr., a Bell Labs researcher, published his criterion in 1956 as a solution to a problem in information theory: given a noisy communication channel with known signal and noise characteristics, what bet size maximizes the long-term growth rate of capital? The same mathematics apply directly to sports betting, where the "signal" is the bettor's edge and the "noise" is the variance of individual outcomes.

The full Kelly formula is f* = (bp - q) / b, where f* is the fraction of bankroll to bet, b is the net decimal odds minus one (the profit per dollar wagered on a win), p is the estimated true probability of winning, and q is the probability of losing (1 - p). When b = 1 (even money, decimal odds of 2.00), this simplifies to f* = 2p - 1, or the edge itself.

For a bet at +150 American (decimal 2.50) that the bettor estimates has a 50% chance of winning: b = 1.50, p = 0.50, q = 0.50. The Kelly stake is (1.50 * 0.50 - 0.50) / 1.50 = (0.75 - 0.50) / 1.50 = 0.25 / 1.50 = 0.1667, or 16.67% of bankroll. A $10,000 bankroll would place a $1,667 bet. This is the mathematically optimal bet size for maximizing the long-term growth rate of the bankroll, assuming the probability estimate is exactly correct.

Why Full Kelly Is Too Aggressive for Most Bettors

The full Kelly formula produces stakes that most recreational and even professional bettors find uncomfortably large. A 16.67% bet means losing five such bets in a row reduces a $10,000 bankroll to $3,900 -- a 61% drawdown. While the Kelly criterion grows wealth faster than any other bet-sizing strategy in the long run, the short-term variance is extreme enough that most bettors cannot withstand the psychological pressure of the drawdowns.

More importantly, the Kelly formula assumes the bettor's probability estimates are perfectly accurate. In practice, every probability estimate contains error. A bettor who estimates p = 0.60 when the true probability is 0.55 will over-bet by a factor of nearly two, because the calculation treats the estimate as ground truth. Overconfidence combined with full Kelly is the single fastest path to bankruptcy in sports betting.

This is why professional bettors almost universally use fractional Kelly -- typically half Kelly (f/2) or quarter Kelly (f/4). Fractional Kelly reduces stake sizes proportionally, cutting variance without sacrificing the mathematical framework that makes Kelly optimal. A quarter Kelly bet on the example above would be $417 instead of $1,667, cutting the maximum drawdown from 61% to approximately 18% while preserving roughly 75% of the long-term growth rate.

Multiple Simultaneous Bets: The Kelly Portfolio Problem

Real bettors rarely place one bet at a time. On a Sunday with a full NFL slate, a serious bettor might have 8 to 12 active wagers simultaneously. When bets overlap in time, the naive approach of applying Kelly independently to each bet over-stakes because it ignores the fact that multiple bets can lose at once, exposing the bankroll to correlated risk.

The correct approach is the Kelly portfolio problem: maximize the expected log growth of the bankroll across a set of simultaneous bets. For small numbers of independent bets, the solution is iterative proportional betting. For correlated bets, the mathematics require covariance estimation and numerical optimization, which is why most professional syndicates use custom software. In practice, the simplest safe heuristic is to reduce total Kelly exposure when placing simultaneous bets: if each of five bets individually calls for 6% of bankroll, capping the sum of all stakes at 15-20% provides a conservative buffer against the variance of multiple outcomes.