I'd like to say that too, but it isn't true. LOL Actually it varies quite a bit from year to year. I've spent years tracking this data...and all I can tell you is that it's pretty inconsistent. Every year is unique it seems. Here's the data I have collected:
Home Dogs:
1999: 46-35, 56.8%
2000: 38-38-2, 50.0%
2001: 42-36, 53.8%
2002: 52-37-2, 58.4%
2003: 36-34-3, 51.4%
2004: 37-42-2, 46.8%
2005: 15-11 thus far, 57.7%
Overall: 266-233-9, 53.3%
Not such a good year last year. In fact, they're basically a 50% play since the start of '03.
I've broken down all the games since '99 by spread ranges and cover rates, home and road covers, etc. The only pattern I can really say I detect in my data is that every season some category (ie. home dogs, road dogs, teams getting 7+ pts, teams laying less than 3, etc) seems to go on a run. But come the next season, throw it all out the window...some other category will be hot, and I have no idea which.
But overall, ya, home dogs would have made you a bit of money if you blindly bet every single one since 1999. (If you pick 52.38% correct you break even taking the juice into account, and since '99 they're a shade over 53%, so they are money-makers overall, but not huge).
I have a theory about why they are under-performing. Simply, there are more people than ever betting on football, and they are more educated than ever. There are still a lot of suckers, but a greater % of the betting public is aware of the value of dogs than was the case years ago. Back in the 80's and early 90's you could probably have made a killing playing them all blindly, but not anymore.