>and I actually watch 48m basketball games, and if you can't
>see who is causing / when runs are occurring, then you
>probably need to study the game more. and anyway once
>playoffs roll around any team moving past the 1st round is
>going to have a recognizable rotation with searchable stats.
>I just believe that will be true from now until they revamp
>how the playoffs work.

like what the fuck is your damn point with this? +/- is a bad one person stat. The 5 man rotations don't have many minutes together: http://www.82games.com/0910/09BOS2.HTM#5man

There are only one or two of those with enough minutes for me to even consider starting to weight them by anything beyond a very basic measure of the strength of opponent, and that's for Rondo, a guy who played 75% of the minutes that season. So yeah, you could weight starting units and maybe one or two other units, but I doubt you could get any other estimation technique to be stable with those small sample sizes.

-----------
It's only funny till someone gets mad. Then it's hilarious.