A bet builder lets you put several selections from the same match into one bet, for example Over 2.5 and both teams to score (BTTSi). It looks like a parlay. It is not.
Correlation
In different matches the events are nearly independent. In the same match they are not. If both teams score, three goals become more likely. The legs push each other.
For independent events the joint probability is the product. For correlated ones it is higher or lower, depending on the link. Here it is higher.
Why the price is not the product
The bookmaker knows about the correlation and builds it into the price. If the single prices are Over 1.85 and BTTS 1.90, the product is 1.85 × 1.90 = 3.52. The combo will be priced much lower, say 2.25. If you expect 3.52 you will be disappointed.
| How it is calculated | Odds | Implied probability |
|---|---|---|
| Product of the single prices | 3.52 | 28.4% |
| Bet builder price (example) | 2.25 | 44.4% |
| Fair price from 42% | 2.38 | 42.0% |
With a 42% probability and a price of 2.25, EV is 0.42 × 2.25 − 1 = −5.5%. The margin here is larger than on a single.
Correlation is not always positive
Some legs from the same match pull in opposite directions. Then the joint probability is lower than the product, and the combo may be priced higher than the product of the single prices.
What to ask before you play
- Are the legs linked? If so, the price is not a product.
- What price would I get if I played each leg separately?
- Do I have my own estimate of the joint probability, not only of each leg?
- What happens if the match or a player does not finish? Read the terms.