[illfuuptn]

I'm going to compile data for an MLB model but injuries pose a bit of a problem. If I just test my model against opening lines, it should do well since my model will have the game time lineups in place already. i.e. the Yankees announce that A-Rod will not play and the line moves 10 points. My model makes me look like a genius because I picked the other team and then line moved 10 points in my direction. Where can I find an archived database of that information?

P.S. Note that this IS extremely different from archived injuries in general. It's easy to find out that "David Ortiz missed game x vs team y, but I need to know when it was announced that he would miss that game.

[jgilmartin]

SportsOptions has an archive of their critical notices, which include a timestamp. I can't remember if Don Best archives them also; they might. You could get a free trial if you haven't done that already.

[illfuuptn]

SportsOptions has an archive of their critical notices, which include a timestamp. I can't remember if Don Best archives them also; they might. You could get a free trial if you haven't done that already.

They do? I just thought they had archived odds and scores. I know they have xml feeds for injuries, but archived? I'll have to check it out. Thanks for the heads up. I'll throw some points your way if you're correct.

NM You said OPTIONS not INSIGHTS my mistake

[illfuuptn]

i'd just use the actual lineups in the game and test against the closing lines.

Any thoughts on this? I like the idea, but will it distort my model too much? I would assume that I should still beat the no-big closing line(if I'm +EV) since the lines don't get moved all the way to perfect equilibrium. So I guess this should still be a sufficient grading system.