Search found 859 matches

by DSMok1
Tue Mar 26, 2019 1:40 pm
Forum: APBRmetrics
Topic: Best practices for SPM development?
Replies: 2
Views: 63

Re: Best practices for SPM development?

At a basic level, developing a SPM model just involves regressing some player stats against a RAPM sample, but what is the best way to do that? Is it: 1. Regress a multi-year player sample against a multi-year RAPM sample (i.e. Kevin Durant's 2010-2015 box score stats against Durant's 2010-2015 RAP...
by DSMok1
Fri Mar 22, 2019 4:18 pm
Forum: APBRmetrics
Topic: 2019 Draft
Replies: 5
Views: 248

Re: 2019 Draft

A rough draft model, based on BPM with modification based on Assists and Usage (positive) and blocks (negative), plus an adjustment for class, gives the following: Year Rk Rk Player Class Season Pos School OBPM DBPM BPM Draft BPM 1 4 Kyrie Irving FR 2010-11G Duke 10.8 3.5 14.3 0.2 2 22 Javon McCrea ...
by DSMok1
Fri Dec 28, 2018 5:46 pm
Forum: APBRmetrics
Topic: 2018-19 season commentary
Replies: 390
Views: 157618

Re: 2018-19 season commentary

I don't want to take credit for shadow's data; They linked their own data set. My 5 year RAPM (2014-15 - 2018-19, regular season only) can be found here https://drive.google.com/file/d/1OvFh3c_NguxDaHOtmLmWnJpOTo3O32SR/view?usp=sharing This is excellent work! Can you please include the number of po...
by DSMok1
Fri Dec 28, 2018 5:13 pm
Forum: APBRmetrics
Topic: Stability of Team Statistics
Replies: 2
Views: 3925

Re: Stability of Team Statistics

A thought experiment: How much time is required for a coach to know if a particular lineup is working well? I know Crow has mentioned often that lineups are used for such short stints, it is hard to know how well a lineup meshes. Plus/minus stats on even a few hundred minutes are extremely unreliabl...
by DSMok1
Fri Dec 21, 2018 3:45 pm
Forum: APBRmetrics
Topic: Notes about RPM estimates
Replies: 13
Views: 8532

Re: Notes about RPM estimates

RPM, as shown on the ESPN website, does not include data from prior seasons AFAIK. Jerry does have a multiyear model for projections.

The RPM box score prior is quite strong early in the season, and that prior does not necessarily match Box Plus/Minus or any other box score metric.
by DSMok1
Fri Dec 21, 2018 3:14 pm
Forum: APBRmetrics
Topic: WOWY Shot charts - http://nbashotcharts.com
Replies: 3
Views: 4410

Re: WOWY Shot charts - http://nbashotcharts.com

Rd11490 wrote:
Fri Dec 21, 2018 3:05 pm
I've added RAPM, Luck Adjusted RAPM, and Real Adjusted Four Factors to the site: http://nbashotcharts.com/fourfactors

The underlying api is open to the public, so if you want to use the data directly, feel free to reach out and i'll send you the request/response model.
Excellent work!!
by DSMok1
Thu Dec 20, 2018 1:04 pm
Forum: APBRmetrics
Topic: Notes about RPM estimates
Replies: 13
Views: 8532

Re: Notes about RPM estimates

Among other top contenders, Raptors have 2 at -2 or worse. Celtics 1, Bucks 0, Warriors 2, Nuggets 1, Sixers 0. They may play moderate to lower minutes and likely less in playoffs. But having a lot of these guys and almost all in regular season rotation does affect the Thunder. This is just a guess...
by DSMok1
Tue Dec 18, 2018 8:12 pm
Forum: APBRmetrics
Topic: STICKY--Registration Problems
Replies: 2
Views: 5601

STICKY--Registration Problems

There seem to be intermittent email system problems with this forum.

If you are having issues registering, please email me directly at my forum username @ hotmail.com.

Thanks, and we are working to resolve the issues.
by DSMok1
Tue Dec 18, 2018 5:25 pm
Forum: APBRmetrics
Topic: Stability of Team Statistics
Replies: 2
Views: 3925

Stability of Team Statistics

Kostya Medvedovsky did some excellent work analyzing the stability of NBA team stats. This would also apply to lineup-level stats. Here are the string of Tweets: https://twitter.com/kmedved/status/1063456469511737344 https://twitter.com/kmedved/status/1063457095926853633 https://twitter.com/kmedved/...
by DSMok1
Mon Nov 12, 2018 2:52 pm
Forum: APBRmetrics
Topic: Regressing toward a prior: Model Selection Question
Replies: 3
Views: 4317

Re: Regressing toward a prior: Model Selection Question

Interesting discussion! To me, this looks like an excellent topic to view, as you suggest, through as Bayesian framework. In other words, we develop a prior expectation entering the year, with an associated error term, and then update with the current year's data (with its associated error term). Th...
by DSMok1
Mon Nov 05, 2018 9:19 pm
Forum: APBRmetrics
Topic: Is there any open source RAPM code?
Replies: 2
Views: 3627

Re: Is there any open source RAPM code?

There are a few codebases floating around for RAPM. One recently posted was here: viewtopic.php?f=2&t=9491
by DSMok1
Wed Oct 17, 2018 11:40 am
Forum: APBRmetrics
Topic: 2018-19 Individual RPM Projections
Replies: 8
Views: 6949

Re: 2018-19 Individual RPM Projections

Jokic is number 1?? How did he get such a high DRPM?
by DSMok1
Sat Oct 13, 2018 11:19 am
Forum: APBRmetrics
Topic: Bayesian RAPM
Replies: 21
Views: 13299

Re: Bayesian RAPM

The idea strikes me as Bayesian at a conceptual level, and I trust it gives you better results than not doing it, but this isn't really a prior. Like the name says, your prior is supposed to be your incoming information before you collect/add data. Priors generally also have a range of possibilitie...
by DSMok1
Fri Oct 12, 2018 12:39 pm
Forum: APBRmetrics
Topic: Bayesian RAPM
Replies: 21
Views: 13299

Re: Bayesian RAPM

My main ones have to do with how the prior is implemented. You said that it varies across seasons, like for Kobe. When you use MPG and team strength, is that from the prior season or the current season? E.g., for a stint from game 50 in 2016, is the MPG/team strength from 2015? From 2016 but only g...
by DSMok1
Thu Oct 11, 2018 2:20 pm
Forum: APBRmetrics
Topic: Bayesian RAPM
Replies: 21
Views: 13299

Re: Bayesian RAPM

xkonk wrote:
Thu Oct 11, 2018 12:29 am
I have lots of questions, but not sure how much Daniel wants to get into it.
I'll try to answer the best I can! I'm not exactly an expert on the mathematical nuances, though.