I am posting here a line that is way off of my model line. Most of the days I find a line so off that I believe my model is broken. But it still give the correct line in 80% of the games, so there has to be another explanation.
For example, yesterday LAD - CWS. The line was +144 / -154, while my model said it was a pick'em game. Both teams have about the same power rating; Lilly has been a solid starter for years, while Humber is an untested starter (40 unremarkable IP in MLB last 4 years, converted to starter this year with mixed success in 45 IP). So, the starter advantage more or less compensated the HFA. So, why was White Sox so heavy favorite?
The WTF line of today is TB - FLA. In team power ratings TB has a very small advantage (.530 vs .527). The issue is the starter. Price is a quality starter, while Vazquez is projected as average. I guess the line is influenced by Vazquez bad numbers this season, but 37 IP should not weigh so heavily in the big picture. Besides, most of the advantage should be neutralized by the HFA, so my model says TB is barely favored. But the line is -156 / +147. Could someone explain?