Help Calculating APM Using R

[CrackersPhinn]

I recently learned a bit about how to use R and I am looking for something to do with it. I thought calculating APM for the 2016-17 might be a reasonable project for me this summer.

I was wondering if people on this forum could give me some advice on how to proceed. I think the most important part would be advice on the data side (i.e. where would be the best place to find pbp data) but I would also appreciate any more general advice experienced stat-creators might have for me.

Thanks a lot.

[Crow]

This request has come up before. There should be some advice in past RPM, RAPM and APM threads. I didn't find the links immediately but there is advice around. I think someone mentioned something on reddit. Or maybe there is a guide by Evan Z. Other guides for other softwares.

[CrackersPhinn]

Based on my initial perusal, it looks like there is no way to get pbp data unless you want to pay for it. Does that sound about right to you?

[DSMok1]

Based on my initial perusal, it looks like there is no way to get pbp data unless you want to pay for it. Does that sound about right to you?

No, you can scrape it and parse it off of the NBA website or other websites. It's not particularly easy, though. The most difficult part is working out who is on the court at any given time, since that is not included in play-by-play data.

I believe that most of the people that do adjusted plus minus have also worked out how to scrape and parse play by play data. That is the more difficult side of adjusted plus-minus than actually setting up and doing the regression.

[Jinxed]

In.

I plan on making this a fun summer project of mine as well.