The discussion of the analysis of basketball through objective evidence, especially basketball statistics.
It is currently Wed Jun 20, 2018 3:16 pm

All times are UTC

Post new topic Reply to topic  [ 228 posts ]  Go to page Previous  1 ... 12, 13, 14, 15, 16
Author Message
PostPosted: Sun Jun 10, 2018 10:36 pm 

Joined: Thu Apr 14, 2011 11:10 pm
Posts: 5184
By several boxscore metrics, Brown and Tatum were pretty average in their large playoff roles. RAPM estimate nice for Tatum, barely above neutral for Brown. Both shot and scored pretty well but neither was that strong at rebounding or assists. Takeaways were alright but not high.

Horford lead the way, as he should for that salary. But what about next season?

Past the first 6, all the rest of Celtics had weak raw team plus minus. Assembled on the cheap. Need better.

PostPosted: Mon Jun 11, 2018 11:39 am 

Joined: Fri Apr 15, 2011 12:02 am
Posts: 4125
Location: Asheville, NC
shadow wrote:
Playoff RAPM updated through game 3 of Finals:

https://docs.google.com/spreadsheets/d/ ... ingle=true
It seems to be updated thru the end of the Finals. Good job!
The Cavs won 12 of 22 games but only 9.7 in Pythagorean wins. And they total just 8.4 in these RAPM Wins. How does that work?
Dubs total not quite 13, instead of 16 or 16.6

Is it unusual that no player has negative RAPM Wins?
The spread of RAPM is pretty tight, with only Rodney Hood and Luc Mbah a Moute below -2; and even they have positive R Wins.

A close fit with the R Wins column is made by the formula:
RW = Min*(RAPM+2.28)/1540

The Cavs are still topsy-turvy in correlation between Min. and D-RAPM. LeBron is their worst defensive player until #10 Rodney Hood. Tristan and JR are next-worst, while Korver is best?

PostPosted: Mon Jun 11, 2018 6:37 pm 

Joined: Fri May 29, 2015 3:38 am
Posts: 189
I'm not exactly sure how JE has setup his playoff RAPM calculations, but I kept the lambda value at 3000 (which I believe is the typical value used in regular season RAPM) when I ran it for the playoffs. That is what's contributing to the tighter spread in ratings and slightly lower than expected win totals, since even the players on the Cavs and Warriors only played around 25% of the minutes they would play in a full regular season. I suppose I could try running it with a lambda of say 750 (3000/4), but that may lead to some pretty low minute players shooting way up the ratings, since nearly half of the players played only 4-7 games at the most and some had even fewer if they only played in select first round games.


I was using too high of a value for replacement level in my regular season RAPM wins formula and had copied it over to the playoff sheet. I had it set to -1.65 ORAPM and -0.6 DRAPM for some reason. According to the link below, I should be using -2.65 for ORAPM and -0.6 for DRAPM. The RAPM wins numbers per team match up a little better with the pythag wins now. I also read that the method I use for calculating RAPM wins isn't going to match up perfectly with pythag wins because my method assumes wins scale linearly with point differential, whereas the pythag method applies diminishing returns at the extremes.

http://godismyjudgeok.com/DStats/2012/n ... ent-level/

Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 228 posts ]  Go to page Previous  1 ... 12, 13, 14, 15, 16

All times are UTC

Who is online

Users browsing this forum: No registered users and 20 guests

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group