APBRmetrics

A team has many levels & parts, but... if you want to look at GMs and give them credit / blame for everything on their watch, you could do GM adjusted plus - minus right? I think it would be an interesting side project.

Wouldn't it be similar to the coaching RAPM that was made a few years ago?

Conceptually yes. But the change dates would vary. Potentially you could do two "actor" GM-Coach model.

I'd think you would have too. I think GM P-M could be too far out there to really have much value. Even the coaching RAPM wasn't all that useful. I think there are too many outside factors to really be able to attribute actions on the court to coaches/GMs.

This is a common debate question. Is it better to do it look at it, think about it and then do other stuff and think some more or proceed without even trying it? I prefer to try stuff but it seems some folks would rather not if something looks like it would be shaky. Less data over clutter or more confusion. To each their own way.

If I had a way of doing RAPM set up I'd try it, but I don't. I'm just not sure GM is close enough to action on the court for the potential results to mean anything.

Almost impossible because of the lack of data. Any work on this would likely be statistically insignificant. Results would probably lie between the residuals.