Go take a look at Ed Gunther’s post over at The National Championship Issue about how all D-1 teams have done against the spread during the period of 2000-8. The data is exhaustive and a fun read, but the must read part is at the beginning, where he reminds us what the spread is about:
… The first thing to realize, if you don’t already, is that the spread is NOT a measure of how much Vegas thinks a team is going to win or lose by. It is a measure of what Vegas thinks the betting public thinks a team is going to win or lose by. Their goal is to get the number of bets on either side of the line as close to equal as they can, thereby ensuring that they’ll win as much as they lose, making a profit on the vig they charge to place the bets. So it’s in their best interest to know not only how good teams are and how they match up, but more importantly what the betting public thinks of how good teams are and how they match up.
So the spread can be seen as the mean, or middle, of public expectation. Half the people think a team will beat the spread, the other half don’t. Looking at how teams have performed against these expectations can give us a good idea into whether they’re truly underrated, overrated, or rated just about right.