[MCS]

Hmmm. Interesting. I don't have a big enough sample size either. I used it largely for ESPN.com games where you have to pick winners and assign confidence levels, and it did very well there. However, I suppose it's possible that it was off on like one game a week, but that I would never have known because that's just not a lot of data.

I always paid more attention to college basketball than college football. I suspect there is less variance in basketball scores; it's a lot more common to have misleading score margins in football. Also, there may be more data in basketball or the teams may be more connected.

It's interesting that you found so many games that were off. I'll have to start looking at more games. Because I really haven't seen what you have, but I assume that you looked at many more.