Kelly Criterion Calculator
Find the optimal bet size for your edge — the stake that maximizes long-run bankroll growth, with safer half- and quarter-Kelly options.
e.g. +150 or -110. Use your true (no-vig) win probability, not the book’s implied odds.
Most bettors use half or quarter Kelly to cut variance. Percentages are of bankroll.
Omenizer sizes every opportunity with a Kelly-based stake — and shows how similar bets actually performed.
See live value bets →The Kelly formula
f* = (b·p − q) / b b = decimal odds − 1 (net fractional odds) p = your win probability q = 1 − p half Kelly = f*/2 · quarter Kelly = f*/4
Worked example
You can bet +150 (decimal 2.50) on a team you rate a true 45% to win. Here b = 1.5, p = 0.45, q = 0.55, so full Kelly = (1.5 × 0.45 − 0.55) ÷ 1.5 = 8.3% of bankroll (the bet’s EV is +12.5%). On a $1,000 bankroll that’s $83 full Kelly, $42 half, or $21 quarter — most bettors take the quarter to smooth the ride.
Kelly stake by win rate (at −110)
How the optimal stake grows as your true win rate climbs above the −110 break-even (52.4%).
| True win rate | EV | Full Kelly | ¼ Kelly |
|---|---|---|---|
| 52.4% | 0.0% | 0.0% | 0.0% |
| 54% | +3.1% | 3.4% | 0.9% |
| 56% | +6.9% | 7.6% | 1.9% |
| 58% | +10.7% | 11.8% | 2.9% |
| 60% | +14.5% | 16.0% | 4.0% |
| 65% | +24.1% | 26.5% | 6.6% |
FAQ
- What is the Kelly Criterion?
- A formula that gives the bet size which maximizes the long-run growth rate of your bankroll, given your edge. Bet too much and you risk ruin; too little and you leave growth on the table.
- Should I use full Kelly?
- Most bettors use a fraction — half or quarter Kelly — because full Kelly is volatile and any error in your probability estimate is amplified. Fractional Kelly keeps most of the growth with far less swing.
- What if the Kelly stake is zero or negative?
- That means the bet has no edge at your estimated probability — the formula says don’t bet. A negative result means the price is worse than fair.
- What win probability should I use?
- Your best honest estimate of the true probability — ideally the no-vig fair probability from a sharp market, not the bookmaker’s implied probability (which includes their margin).
- Can I still lose money using Kelly?
- Absolutely. Kelly maximizes long-run growth, but variance is real — even a bankroll of +EV bets goes through losing streaks. Fractional Kelly and a large bet sample are how you survive them.
- Why is overbetting so dangerous?
- Kelly is very sensitive to your probability estimate. If your edge is smaller than you think, full Kelly overbets and can shrink your bankroll even on +EV bets. Betting a fraction protects against estimate error.
- Kelly vs. flat staking — which is better?
- Flat staking (same amount each bet) is simpler and lower-variance; Kelly grows faster when your edges are accurate. Many bettors use quarter or half Kelly as a middle ground between the two.
Related free tools
Built by the team behind Omenizer’s real-time fair-odds engine — the same devigging and closing-line-value math that powers our live value-bet feed. Last updated July 2026.
Educational tool only. Not betting advice or a guarantee of profit. Kelly assumes your probability estimate is accurate.