Common Kelly Criterion Mistakes
📖 Why Does the Kelly Criterion Fail?
The Kelly Criterion is a powerful mathematical tool, but many bettors make mistakes that compromise their results. Knowing these mistakes is the first step to using it correctly.
Mistake #1: Using Full Kelly (100%)
The most common mistake is using Full Kelly, which is extremely aggressive and can lead to rapid ruin.
$1000 bankroll, odds 2.0, 60% real probability. Full Kelly suggests betting 20% ($200). A sequence of 3 losses breaks the bankroll.
Use Fractional Kelly (25% or 50%). The same example with Kelly 25% suggests only 5% ($50), drastically reducing risk.
Mistake #2: Overestimating Probabilities
Bettors tend to overestimate their chances, especially with teams they support or events they follow.
Odds 2.0, you estimate 70% chance (Kelly suggests 40% of bankroll), but the real chance is 55% (Kelly would suggest 10%). You bet 4x more than ideal.
Be conservative with your estimates. Use external sources (odds from other bookmakers, statistical models) to validate your perception.
Mistake #3: Ignoring Bankroll Size
Kelly suggests betting a percentage, but with small bankrolls this can be problematic.
$100 bankroll, Kelly suggests 10% ($10). A $10 bet represents high risk for a small bankroll.
Set a maximum bet limit (e.g., never more than 5% of bankroll) and use fractional Kelly.
Mistake #4: Betting Without Mathematical Edge
Many bettors use Kelly even when they have no edge over the odds.
Odds 2.0, you estimate 50% chance (equal to odds). Kelly suggests 0%, but you bet based on "intuition".
Only bet when you have a mathematical edge. Calculate the expected value before each bet.
Expected Value = (Probability × Odds) - 1
Only bet if Expected Value > 0
Mistake #5: Not Recalculating Bankroll
Kelly requires the bet size to be recalculated each time, based on the current bankroll.
You calculated 10% of $1000 bankroll ($100). After losing, your bankroll dropped to $900, but you continue betting $100.
Recalculate the bet size after each result, based on the current bankroll.
Mistake #6: Not Setting a Stop-Loss
Even with Kelly, you can have bad sequences. Without a stop-loss, you could lose your entire bankroll.
A sequence of 5 consecutive losses with Kelly 25% can reduce your bankroll by ~50%.
Set a daily stop-loss (e.g., stop after losing 20% of bankroll) and a per-bet stop-loss (e.g., max 5% of bankroll).
📊 Summary: Mistakes and Solutions
| Mistake | Consequence | Solution |
|---|---|---|
| Full Kelly (100%) | High risk of ruin | Use Kelly 25% or 50% |
| Overestimating probabilities | Betting too much above ideal | Be conservative, validate with external sources |
| Ignoring bankroll size | Disproportionate risk | Maximum 5% of bankroll limit |
| Betting without edge | Guaranteed long-term loss | Only bet with positive expected value |
| Not recalculating bankroll | Risk imbalance | Recalculate after each bet |
| No stop-loss | Total loss in bad sequences | Set daily and per-bet limits |
✅ Pre-Bet Checklist
🎯 Conclusion
The Kelly Criterion is a powerful tool, but like any tool, it needs to be used correctly. Avoid the common mistakes listed above, be conservative with your estimates, and always use Fractional Kelly.
Remember: The goal is not to maximize profits at any cost, but to grow sustainably over the long term.