GAME 2: UTA (1-1) vs. DAL (1-1)| 104-110 WIN!!!!!

[mvossman]

Totally off topic here, but I've never gotten myself to the place where I could like baked plus minus stats very much. I've read the papers put out by the creators of many of the metrics and just don't come away convinced by the math or by the results. 

So far, imo, the search for the Grand Theory of Basketball has produced iffy results. Which makes sense to me considering that we're having a really hard time marrying big physics and little physics even though we've been trying for centuries. 

I'd love to hear an expert in probability and statistics weigh in on whether these regression and other exercises have produced anything valuable. 

@"Hogmelon", I can't remember, did you study math, or are you mostly trained via gambling?

So I don't know if this is totally answering your question, and I think I posted this in the past, but here is a really good article going over the various stats and a survey of what basketball folks think of them. 

EPM is the highest rated one (DARKO is something different) and article includes a link to really well done study comparing the effectiveness of the stats.  When Kam posted some of these stats in a thread he started a couple of days ago, the general consensus was that EPM was the one that met the eye test.  I would argue it is even more useful when looking over a larger sample (multiple years).  If there is one stat that you could semi rely on for a catch all, it would probably be EPM.

What is the best advanced statistic for basketball? NBA executives weigh in | HoopsHype