My hunch was right. The 32 teams together win total lines come to 262.5. The correct number would be 8 wins per team or 256 wins total. So the win totals as a whole are mispriced by 6.5 wins.
The moneylines also tell a story. For the overs only two teams, Miami (8, +110o) and Carolina (8.5, +120o) on the over are offering a premium to take the over. The combined over moneylines are -3335, for every $100 wagered the book is only paying $81.20 to winners due to discounting.
The moneyline on the unders looks much better. The combined under moneylines are -325 for every $100 wagered the book is paying $98.15 to winners.
the mystery of the 6.5 wins
as we see there must be more public action on the overs. so the book compensates by adding wins to the lines and by discounting the moneyline. it seems it would be tough to find much value on the overs.
to there is probably more value then in the unders on the win totals. sow how to approach these 6.5 wins that were added to the lines. a first guess is that there are marquee teams with a large fan base who always get a lot of action on the overs. so potential plays for the under could be
| Team | 2014 Win Line | Moneyline (under) |
|---|---|---|
| Dallas Cowboys | 7.5 | -115 |
| Washington Redskins | 7.5 | EVEN |
| Green Bay Packers | 10.5 | -115 |
| San Francisco 49ers | 10.5 | +105 |
| Pittsburgh Steelers | 8.5 | +120 |
| New England Patriots | 11 | -115 |
These teams all have at least some questions at quarterback due to age or significant injuries in 2013. As a group I might not expect to go 6-0 on the unders but I think 3.5-4 wins is doable. they are priced to be favourable to the unders, probably to try to balance the public action on the overs.
a black box approach
now we don't know for sure where the extra 6.5 wins are distributed in the lines. we just know for certain the aggregate winlines are off by 6.5 wins. it should total 256 but the total is 262.5
it makes me wonder if there might be value in just taking the under on all 32 teams to attempt to profit from the pricing error. in general due to the error you would expect to do better than 16-16. at even money this should be a profitable play. in this case there is a moneyline hindrance that winners have a discount where in aggregate only $98.15 is paid to $100.
I realized that with some assumptions it's possible to create a model and run some simulations and calculate the effect of the 6.5 win pricing error and estimate the value of just taking the under on all 32 teams while accounting for the moneyline discount. there's a bit to it so I'll leave it for a future post if I get to it.
No comments:
Post a Comment