It's well known the betting public prefers favorites to underdogs, and in that case you'd think there would be a slight advantage to picking them - because Vegas would have the incentive to move the spreads ever-so-slightly to favor dogs. And you'd be right, though the margin is narrowing.
According to Sports Insights, 51.1 percent of dogs have come in over the last six years, but even more came in before that:
| Year | Win|PERCENT| |
| 2001 | 53.00|PERCENT| |
| 2002 | 55.80|PERCENT| |
| 2003 | 53.20|PERCENT| |
| 2004 | 50.00|PERCENT| |
| 2005 | 42.80|PERCENT| |
| 2006 | 55.90|PERCENT| |
| 2007 | 49.80|PERCENT| |
| 2008 | 50.60|PERCENT| |
| 2003 - 2008 | 51.10|PERCENT| |
Moreover, they tracked what happened when you simply go against the public (when 75 percent or more of the public is on the other side), and also when that pick was a dog:
| Year | All Dogs | Dogs, Bet Against Public |
| 2003 | 53.20|PERCENT| | 61.80|PERCENT| |
| 2004 | 50.00|PERCENT| | 50.00|PERCENT| |
| 2005 | 42.80|PERCENT| | 48.10|PERCENT| |
| 2006 | 55.90|PERCENT| | 69.80|PERCENT| |
| 2007 | 49.80|PERCENT| | 53.70|PERCENT| |
| 2008 | 50.60|PERCENT| | 42.10|PERCENT| |
| 2003 - 2008 | 51.10|PERCENT| | 53.30|PERCENT| |
The issue here is where to find a good source of public information - I use the Sportsbook.com betting trends, but I'm not sure that's a big enough or representative sample of the public, and by the criteria above it virtually ensures that you take every single double-digit underdog all year, every year. (Not that there's anything wrong with that in and of itself, but I don't think the 13-point games between a bad road dog and a tough home favorite are where you make your money. Those strike me more as toss-ups.)
Finally, they track line movement - when the spread moves half a point to offset a big bet on the other side:
| Year | Following Small Line Movement | ||
| 2003 | 61.40|PERCENT| | ||
| 2004 | 58.20|PERCENT| | ||
| 2005 | 58.80|PERCENT| | ||
| 2006 | 51.60|PERCENT| | ||
| 2007 | 51.50|PERCENT| | ||
| 2008 | 47.50|PERCENT| | ||
| 2003 - 2008 | 54.80|PERCENT| | ||
I assume they mean following the line movement, i.e, betting on the worse line after it moved, as an indicator that sharps are confident enough in the team to place big bets on it. If it means how the results came out when looking at the line before it moved, that wouldn't help you much because you can only know after it changes. And I suppose the theory here is that the endorsement of a pro risking his own money (very different than one from a tout trying to get your money) outweighs the half-point move against you.
Also, keep in mind that some years, this stuff works terribly, and so even if you play these angles, and they're right over time, there's no guarantee they hold up in 2009, let alone in a given week or on a given game.