[mathdotcom]

Which game is most likely to be informative, by halftime, about who will eventually win.

A low scoring game with a team favored by 3 points, 7points?
A game where there is a huge dog?

I like the huge dog option. I feel by halftime you have a good idea of whether the Colts are going to cover 14 points or not against the Bills, whereas the 3 point games are often quite close at halftime.

Thoughts gentlemen?

[ronald]

Matthew,

Which books offer this?

[Justin7]

Can you play MLs on it?

[mathdotcom]

Just spreads

[Justin7]

That's a very interesting problem. Is it a 1-shot opportunity? Or can you keep doing it each week, or for every game?

[Data]

The lower the posted total, the lower the variance. The lower the variance, the higher the correlation between 1H and game results.

[mathdotcom]

Thank you sir

Holding the total constant is there any reason to prefer a 7 pt spread to a 3 pt spread?

[Data]

I think the bigger the spread, the higher the probability of the better team having a lack of motivation in either half resulting in higher variance making 1H results the lesser predictors. This hypothesis can be easily checked against a database that I do not posses due to my lack of interest in betting NFL.

[mathdotcom]

That makes two of us Data

Thank you sir