What is Kelly Criterion? A Sports Bettor's Guide to Optimal Sizing

If you have ever placed a bet on Kalshi or any prediction market, you have faced the same question every trader faces: how much should I risk on this position? Bet too little and you leave money on the table. Bet too much and a single loss can wipe out your bankroll. The Kelly Criterion is the mathematically proven answer to this problem, and understanding it is the single biggest upgrade most sports bettors can make to their strategy.

The Problem with Flat Betting

Most beginners use flat betting -- risking the same dollar amount on every trade regardless of how confident they are. While simple, this approach ignores critical information. A contract where you believe the true probability is 70% but the market prices it at 50% deserves a much larger position than one where you see only a 2% edge. Flat betting treats both the same, which is mathematically suboptimal.

The Kelly Criterion Formula

The Kelly Criterion was developed by John L. Kelly Jr. at Bell Labs in 1956. Originally designed for information theory, it turns out to be the optimal strategy for maximizing long-term bankroll growth. The formula is elegantly simple:

f* = (bp - q) / b

Where:

• f* = the fraction of your bankroll to wager
• b = the net odds received on the bet (decimal odds minus 1)
• p = the probability of winning
• q = the probability of losing (1 - p)

For binary contracts on Kalshi, this simplifies even further. If a contract is trading at a price that implies one probability, and your model says the true probability is different, you can calculate the exact percentage of your bankroll to allocate.

A Practical Kalshi Example

Let us walk through a real-world scenario. Suppose you are looking at a Kalshi contract for an NBA game: "Will the Lakers win tonight?" The contract is trading at $0.40, which implies the market believes there is a 40% chance the Lakers win.

Your live win probability data, however, suggests the Lakers actually have a 55% chance of winning. You have identified an edge. But how much should you bet?

Using the Kelly formula:

• Market price (implied probability): 40% ($0.40 per contract)
• Your estimated true probability (p): 55%
• Probability of losing (q): 45%
• Net odds (b): If you buy at $0.40 and win, you receive $1.00, so your net profit is $0.60 on a $0.40 risk. That means b = 0.60 / 0.40 = 1.5

f* = (1.5 x 0.55 - 0.45) / 1.5 = (0.825 - 0.45) / 1.5 = 0.375 / 1.5 = 0.25

Kelly says you should bet 25% of your bankroll on this trade. That is a substantial allocation, which makes sense given the large edge (15 percentage points between your model and the market).

Why Full Kelly is Dangerous

In practice, almost nobody uses full Kelly sizing. The formula assumes your probability estimate is perfectly accurate, which it never is. If your model is even slightly off, full Kelly can lead to massive drawdowns. This is why experienced traders use fractional Kelly -- typically between one-quarter and one-half of the full Kelly recommendation.

Using half-Kelly on our example above, you would bet 12.5% of your bankroll instead of 25%. You sacrifice some theoretical growth rate in exchange for much smoother equity curves and smaller drawdowns.

When Kelly Says "Do Not Bet"

One of the most valuable features of the Kelly Criterion is that it tells you when not to trade. If the formula returns a negative number, it means you have no edge -- or worse, the edge is against you. This is a powerful discipline tool.

For example, if the Lakers contract is at $0.40 but your model says they only have a 35% chance of winning:

f* = (1.5 x 0.35 - 0.65) / 1.5 = (0.525 - 0.65) / 1.5 = -0.125 / 1.5 = -0.083

The negative result means you should not buy this contract. In fact, you might consider selling it (taking the other side) if the negative edge is large enough.

Kelly Criterion and Bankroll Management

Kelly sizing naturally implements bankroll management. When your bankroll grows, Kelly increases your position sizes proportionally. When your bankroll shrinks after losses, Kelly automatically reduces your bets, making it nearly impossible to go bust (in theory, though real-world constraints like minimum bet sizes matter).

This is fundamentally different from fixed-dollar betting, where losing streaks can quickly erode your entire bankroll because your bet sizes do not adjust.

How EventEdge Uses Kelly Criterion

Calculating Kelly sizing by hand for every Kalshi contract is tedious and error-prone. This is one of the core problems EventEdge solves. EventEdge is a Kalshi autotrading bot that continuously compares real-time probability models against live Kalshi market prices. When it detects an edge, it automatically calculates the optimal position size using fractional Kelly Criterion.

Here is what happens under the hood:

1. Edge Detection: EventEdge compares its live win probability data to the current Kalshi contract price to identify mispriced contracts.
2. Kelly Calculation: For every edge it finds, EventEdge runs the Kelly formula using the estimated true probability and the market-implied probability.
3. Fractional Scaling: EventEdge applies a conservative fractional Kelly multiplier to reduce variance and protect against model uncertainty.
4. Automated Execution: If the Kelly-sized position meets minimum thresholds, EventEdge places the trade on Kalshi automatically.

This means you get mathematically optimal position sizing on every single trade, 24/7, without having to do any math yourself.

Common Kelly Criterion Mistakes

Overestimating your edge. The Kelly formula is only as good as your probability estimate. If you think a team has a 60% chance but the true probability is 50%, Kelly will tell you to bet big on a position with no actual edge. This is why using rigorous, data-driven probability models matters so much.

Ignoring correlation. If you have multiple positions that are correlated (e.g., several bets on the same game), you need to account for that. Standard Kelly assumes independent bets.

Using full Kelly. As mentioned, full Kelly is theoretically optimal but practically dangerous. Always use fractional Kelly.

Key Takeaways

The Kelly Criterion is the gold standard for position sizing in prediction markets and sports betting. It tells you exactly how much to bet based on your edge, and equally importantly, when not to bet at all. While the math is straightforward, applying it consistently across hundreds of Kalshi contracts requires automation -- which is precisely what tools like EventEdge are built to do.

If you are serious about generating passive income from Kalshi prediction markets, understanding Kelly Criterion is not optional. It is the difference between gambling and trading with a systematic edge.

---

Want to see Kelly Criterion in action? EventEdge applies fractional Kelly sizing to every trade it makes on Kalshi, automatically.