Player Prop Correlation in Combination Bets

## The Correlation Opportunity in Props Player props are particularly interesting for combination bets because the correlation between some props is strong and predictable — and bookmakers do not always model it accurately. ## Positive Correlations (Both More Likely Together) **Quarterback passing yards + Wide receiver receiving yards:** These are almost perfectly positively correlated in the same game. More team passing yards = more individual receiver yards. A bet builder combining a QB over passing yards with a WR over receiving yards bets the same underlying variable twice — not an independent combination. **Team to score many goals + Leading striker to score:** A team winning 4-2 makes the striker scoring anytime more likely. The combination of "over 3.5 goals" + "striker anytime scorer" is more probable than their individual probabilities multiplied — they are positively correlated. ## Where Bookmakers Under-Price the Correlation Some bookmakers' same-game multi (bet builder) engines use simplified correlation models. When two props are highly positively correlated but the bookmaker treats them as independent (multiplying probabilities), the combined price is understated — the bet builder offers better than fair value. Finding these correlations requires a player × team probability model that estimates joint probabilities. This is advanced modelling work but can produce persistent prop combination edges. ## Negative Correlations to Avoid Some combinations are negatively correlated: - "Team A to win" + "Team B striker to score anytime" — if Team A wins, it is less likely Team B scored (or vice versa). Combining these reduces the true joint probability below the independent product. Bookmakers may overprice negatively correlated combinations — making them worse value than they appear.