APBRmetrics

You got a slice by SMRE and real close on absolute error. Celebration is mostly self-service here... but congrats!

I'll remember my brief come from behind lead, before final week or two of ball where a lot of teams seem to have atypical performances.

Andrew Johnson said it well last year in an article -- he'd finished among the leaders.
Anyone doing better than Vegas has done something.

EvanZ wrote:Never really feel like I learn much from these predictions since a) the totals are fairly close to each other and b) minutes projections could make a significant difference.

Minute information makes a significant difference. Relatively simple models will beat Vegas retrodictively if they can use minute information.

That said, it seems like you'd want to start by setting up the scoring in a way that allows people to control for minutes rather than coming in with that at the end of the contest.

Looks like 538 has realized their mistake and are adding RPM back into their model. Interesting that they felt it warranted complete removal last year but are weighting it at 2/3 now.