The Kelly Criterion tells you the mathematically optimal bet size for maximum long-term growth. So why do nearly all professional bettors deliberately bet less than the full Kelly amount?
The answer comes down to one word: volatility. Full Kelly is aggressive. Very aggressive. And in the real world, where your edge estimates are never perfectly accurate, full Kelly can lead to stomach-churning swings that most bettors cannot handle, both financially and emotionally.
Full Kelly maximizes your expected growth rate, but it also maximizes your volatility. Here is what that looks like in practice:
That last point is crucial. Full Kelly assumes you know your true win probability with perfect accuracy. But nobody does. If you think you have a 55% edge but your actual edge is 52%, you are betting nearly twice as much as you should. This overexposure compounds over time and can lead to significant losses.
Bill Benter, who made many millions using a fractional Kelly approach to racetrack wagering, has noted that even the best computer handicapping models can overestimate the edge by a factor of two. Someone attempting to place a Kelly bet might unintentionally be placing a twice-Kelly bet, which dramatically cuts the compound return rate.
Fractional Kelly means betting a fraction of what the full Kelly formula recommends. The most common fractions are:
Half Kelly is particularly appealing because it sits on the flat part of the Kelly growth curve. The mathematical relationship between bet size and growth rate is a parabola, not a straight line, which means that reducing your bet size from the peak costs you disproportionately little growth relative to how much volatility you shed.
Here is how different Kelly fractions compare for a bettor with a 54% win rate on -110 odds:
| Strategy | Bet Size | Growth Rate | Volatility | Max Drawdown Risk |
|---|---|---|---|---|
| Full Kelly | ~5.5% | 100% | High | Moderate to High |
| Half Kelly | ~2.75% | 75% | Low | Low |
| Quarter Kelly | ~1.4% | ~50% | Very Low | Very Low |
The key insight: halving your bet size sacrifices only about a quarter of your long-term compound growth rate while dramatically taming the swings. That asymmetry, losing a little growth but shedding a lot of volatility, is what makes fractional Kelly so powerful in practice.
Half Kelly has become the industry standard for several reasons:
Even the best handicappers overestimate their edge. Half Kelly provides a buffer. If your true edge is smaller than you think, you are not overexposed. Edward Thorp has explained that it is not primarily uncertainty, but a background tendency to overestimate winning chances, which justifies a partial Kelly strategy.
With half Kelly, losing streaks hurt less and winning streaks still compound nicely. You can stick with the strategy through inevitable rough patches.
Full Kelly can recommend betting 10% or more of your bankroll on a single game. Most bettors cannot handle that emotionally. Half Kelly keeps bet sizes in a range that feels manageable.
Retaining the vast majority of the theoretical growth rate is still excellent. Over hundreds of bets, this compounds into significant returns while letting you sleep at night.
One of the most striking differences between full and fractional Kelly is the probability of large drawdowns. A full Kelly bettor has roughly a 1 in 3 chance of halving their bankroll before doubling it. A half Kelly bettor has only about a 1 in 9 chance of the same.
| Strategy | Chance of 50% Drawdown Before Doubling |
|---|---|
| Full Kelly | ~33% (1 in 3) |
| Half Kelly | ~11% (1 in 9) |
A 50% drawdown with full Kelly means you need to double your bankroll just to get back to even. With half Kelly, you are far less likely to face that situation in the first place.
When to avoid full Kelly entirely: If you cannot verify your edge with hundreds of tracked bets, or if you are betting correlated events on the same day, or if your bankroll is money you cannot afford to lose, stick to quarter Kelly or even lower.
Quarter Kelly is appropriate when:
The best way to understand the difference is to see it visually. You can run a bankroll simulation that compares full Kelly, half Kelly, quarter Kelly, and flat betting over hundreds of bets. Watch how the volatility differs and how often full Kelly experiences deep drawdowns.
The fraction you choose should match your confidence in your edge estimates and your emotional tolerance for drawdowns. There is no single right answer, but the decision framework is straightforward: less data means smaller fractions, and more data means you can afford to size up. Our in-depth fractional Kelly guide walks through the specific criteria for selecting quarter, third, half, or three-quarter Kelly based on your experience level, model sophistication, and correlation between bets.
The one universal rule: when in doubt, go lower. You can always increase your fraction as you build confidence in your edge estimates. Starting too aggressive is a much bigger mistake than starting too conservative.
Full Kelly is mathematically optimal in a world of perfect information. We do not live in that world. Fractional Kelly, especially half Kelly, is the practical choice for real bettors with imperfect edge estimates and finite emotional tolerance for volatility.
The professionals know this. They have seen what full Kelly does during a bad streak. They have watched accounts swing wildly on paper. Almost universally, they have settled on fractional Kelly as the sustainable path to long-term growth.
You can calculate your optimal bet size using any fraction you choose. Start conservative, track your results, and adjust as you validate your edge over time.