Kelly Criterion Calculator | Optimal Bet Sizing Tool
Determine the optimal percentage of your bankroll to wager based on your edge and odds to maximize long-term growth while minimizing ruin risk.
Bet Details
| Implied Probability | -- |
| True Probability | -- |
| Full Kelly (100%) | -- |
| Expected Value (EV) | -- |
| Growth per Bet | -- |
| Exp. Bankroll (after N) | -- |
What is the Kelly Criterion?
The Kelly Criterion is a mathematical formula developed by John Kelly Jr. at Bell Labs in 1956. It determines the optimal amount to bet on a wager with a known edge to maximise long-term wealth.
Unlike flat betting (wagering the same amount every time) or gut-feel betting, Kelly sizes your position proportionally to your advantage. It is widely used by professional investors, poker players, and sports bettors to balance aggressive growth with bankroll preservation.
The Kelly formula explained
f* = (bp − q) / b
- f* = fraction of bankroll to bet
- b = net odds (decimal odds − 1)
- p = probability of winning
- q = probability of losing (1 − p)
If the formula spits out a negative number, the bet has no edge and the optimal Kelly stake is zero — don't bet.
How to use this calculator
- Enter bankroll — optional, but lets you see actual dollar stakes alongside percentages.
- Select odds format — Decimal, American, or Fractional.
- Input odds — the price offered by the sportsbook.
- Estimate your win probability — this is the crucial input. Use historical data, models, or deep research to find your honest estimate of the true win %.
- Set the multiplier — 0.5 (Half Kelly) is the default; quarter or full Kelly are alternatives.
The bar chart on the right compares Full / Half / Quarter Kelly stakes side by side; the line chart projects expected bankroll growth over 0–100 bets at your current settings versus not betting at all.
Worked example: NFL betting
| Step | Value |
|---|---|
| Scenario | Cowboys moneyline at −110 |
| Implied probability | 52.38% (break-even point) |
| Your estimate | 55% (you have an edge) |
| Your edge | 2.62% |
| Recommendation (Half Kelly) | Bet 2.75% of bankroll ($27.50 on $1,000) |
Fractional Kelly guide
| Multiplier | Behaviour | Approx. risk of ruin |
|---|---|---|
| Full Kelly (1.0) | Max growth, high volatility. Only for experts with huge datasets. | ~13% |
| Half Kelly (0.5) | Industry standard. ~75% of max growth with much lower risk. | ~1.8% |
| Quarter Kelly (0.25) | Safety first. Smooth growth curve. | ~0.03% |
Common mistakes to avoid
- Overestimating win probability. The #1 cause of ruin. If you think you have a 60% chance but the true probability is 55%, Kelly will suggest overbetting.
- Ignoring variance. Full Kelly is mathematically optimal for growth but can result in 50%+ drawdowns. Most professionals run a fractional Kelly to smooth the ride.
- Misunderstanding "risk of ruin". The figure here is a theoretical asymptotic probability of a significant drawdown (e.g. losing more than 95% of bankroll), not necessarily the probability of literally hitting $0.
Frequently asked questions
What is the best Kelly multiplier for beginners?
Most pros recommend 0.25 (Quarter Kelly) to 0.5 (Half Kelly). This captures most of the long-run growth while drastically reducing the risk of ruin.
Does Kelly guarantee profits?
No. Kelly optimises stake size for your estimated edge. If your estimate is wrong, you will lose money efficiently. The formula is only as good as the win probability you feed it.
Can I use Kelly on parlays?
Yes, but calculate the combined true probability first (multiply the legs' independent probabilities). Because parlays have very high variance, use Quarter Kelly at most.
Why does the calculator recommend $0?
If your edge is zero or negative — i.e. the sportsbook's implied probability is at or above your estimated true probability — Kelly's optimal bet is always zero. The math is telling you not to bet.