The mathematical formula that tells you exactly how much to bet based on your edge. Used by hedge funds, professional gamblers, and now available in betting research apps.
The Kelly Criterion is a mathematical formula developed by John L. Kelly Jr. at Bell Labs in 1956. Originally designed for information theory (optimizing signal-to-noise ratios in telephone lines), it was quickly adopted by gamblers and investors as the optimal strategy for sizing bets when you have an edge.
The core idea is simple: bet more when you have a bigger edge, and less when your edge is smaller. The formula tells you the exact percentage of your bankroll to wager on each bet to maximize long-term growth while minimizing the risk of going broke.
Edward Thorp, the mathematician who famously beat Las Vegas at blackjack, used the Kelly Criterion as the foundation of his betting strategy. Warren Buffett and Bill Gross have cited it as influential in their investment philosophies. Today, it's the standard framework used by professional sports bettors and quantitative hedge funds.
Where:
You believe the Patriots have a 62% chance of covering -7 at -110 odds (decimal: 1.909).
f* = (0.909 × 0.62 - 0.38) / 0.909 = 0.201 / 0.909 = 0.221
The Kelly Criterion recommends betting 22.1% of your bankroll on this bet. In practice, most bettors use "fractional Kelly" (half or quarter Kelly) to reduce variance.
The formula works the same way regardless of odds format — you just need to convert to decimal odds first:
| American Odds | Decimal Odds | Conversion |
|---|---|---|
| -110 | 1.909 | 100/110 + 1 |
| +150 | 2.500 | 150/100 + 1 |
| -200 | 1.500 | 100/200 + 1 |
| +300 | 4.000 | 300/100 + 1 |
| EVEN (+100) | 2.000 | 100/100 + 1 |
The Kelly Criterion has been mathematically proven to maximize the geometric growth rate of your bankroll over time. No other betting strategy grows your money faster in the long run — assuming your probability estimates are accurate.
One of the biggest mistakes recreational bettors make is betting too much on any single wager. The Kelly Criterion automatically scales your bet size down when your edge is small and up when it's large, preventing catastrophic losses.
If the Kelly formula returns a negative number, it means you don't have an edge — the bet has negative expected value. This is incredibly valuable information that most bettors ignore.
Full Kelly betting is mathematically optimal but practically aggressive. The swings can be stomach-churning. Most professional bettors use fractional Kelly:
| Strategy | Bet Size | Volatility | Best For |
|---|---|---|---|
| Full Kelly | 100% of Kelly | Very High | Theoretical maximum growth |
| Half Kelly | 50% of Kelly | Moderate | Most professional bettors |
| Quarter Kelly | 25% of Kelly | Low | Conservative / learning bettors |
| Eighth Kelly | 12.5% of Kelly | Very Low | Large bankroll preservation |
The Kelly Criterion formula is easy. The hard part is the p variable — accurately estimating the true probability of winning a bet. If your probability estimates are wrong, your Kelly bet sizes will be wrong.
This is where most bettors fail. They either:
Modern AI-powered tools like Juice solve this problem by using artificial intelligence to research games and estimate true probabilities, then automatically applying the Kelly Criterion to recommend optimal bet sizes. Instead of guessing probabilities, the AI analyzes injury reports, historical trends, and matchup data to generate data-driven estimates.
Flat betting (wagering the same amount on every bet) is simpler but suboptimal. Here's how they compare over 1,000 bets with a 55% win rate at -110 odds:
| Strategy | Starting Bankroll | Ending Bankroll | Growth |
|---|---|---|---|
| Flat (2% per bet) | $10,000 | $11,820 | +18.2% |
| Quarter Kelly | $10,000 | $12,450 | +24.5% |
| Half Kelly | $10,000 | $14,200 | +42.0% |
| Full Kelly | $10,000 | $17,800 | +78.0% |
Note: These are theoretical averages. Actual results will vary based on the sequence of wins and losses.
Juice automatically applies the Kelly Criterion to every bet analysis, giving you optimal bet sizing based on AI-estimated probabilities.
Try Juice Free for 3 Days →