Search found 194 matches

by talkingpractice
Fri Oct 24, 2014 5:45 pm
Forum: APBRmetrics
Topic: Predictions 2014-2015
Replies: 265
Views: 290313

Re: Predictions 2014-2015

a lot of gsw's success this year may come down to removing jackson, and hence getting rid of all those hockey substitutions and the "barnes iso thing".

ofc kerr may put iggy on the bench (?), so idunno...
by talkingpractice
Thu Oct 16, 2014 3:17 pm
Forum: APBRmetrics
Topic: Possible Steps to improve SPM
Replies: 55
Views: 31098

Re: Possible Steps to improve SPM

Including the individual shot defense component? Are you using it to improve your spm or IPV? we did find some benefit to using opponents fg% at the rim, but nothing else (for defense). offense doesnt have anything useful yet. but a lot of that is due to having just one year of sport vu data (i thi...
by talkingpractice
Wed Oct 15, 2014 4:40 pm
Forum: APBRmetrics
Topic: Possible Steps to improve SPM
Replies: 55
Views: 31098

Re: Possible Steps to improve SPM

i think we need 1-2 more years of sportvu to have something useful for prediction.

i can already vouch for Vantage data.
by talkingpractice
Sun Sep 07, 2014 10:10 pm
Forum: APBRmetrics
Topic: Predictions 2014-2015
Replies: 265
Views: 290313

Re: Predictions 2014-2015

the gotbuckets RAPMs are a RiRAPM variant that we provide, for people interested in a pure RAPM model. it does include playoffs. youd do better using jerry's xRAPM, probably. you definitely want to eliminate WP from whatever youre doing. if we put on any long positions in these this year, ill probab...
by talkingpractice
Wed Aug 27, 2014 9:57 pm
Forum: APBRmetrics
Topic: ESPN/J.E.: Rookies provide no impact at all
Replies: 23
Views: 13878

Re: ESPN/J.E.: Rookies provide no impact at all

imo, a lot of people have attributed about 234923423 things to je here that he either didnt really say, didnt really mean, or both. all he was really saying is that rookies, almost always, tend to suck. everything else is just sorta conversation of particulars and semantics that don't matter that mu...
by talkingpractice
Mon Aug 25, 2014 5:09 pm
Forum: APBRmetrics
Topic: machine learning draft prediction model
Replies: 10
Views: 5882

Re: machine learning draft prediction model

hey, thanks for coming by and replying. perhaps for dv, youd want to try something RAPM-based? you could still include something for longevity by using MP as you did originally. PER doesnt handle defense well at all, and may overstate the benefit of usage (prolly not as much as a lot of people think...
by talkingpractice
Thu Aug 21, 2014 2:31 am
Forum: APBRmetrics
Topic: machine learning draft prediction model
Replies: 10
Views: 5882

Re: machine learning draft prediction model

Another aspect that might improve this analysis is to try more than one regression technique (AIC, BIC, LASSO, ELASTICNET, RIDGE..), or at least mention which kind was used. If he used, let's say, stepwise regression with t-tests then that would (very likely) lead to worse results than with the met...
by talkingpractice
Wed Aug 20, 2014 12:00 am
Forum: APBRmetrics
Topic: 2014-5 Celtics
Replies: 1
Views: 1081

Re: 2014-5 Celtics

if we lose 50 or more, then we maximize our chances for okafor ;)
by talkingpractice
Tue Aug 19, 2014 11:58 pm
Forum: APBRmetrics
Topic: machine learning draft prediction model
Replies: 10
Views: 5882

Re: machine learning draft prediction model

imo, the big flaw here is that he picked a baddish dv, and no matter what machine learning technique you use, it cant overcome a bad choice of dv.

and i dont get why he tried all the things that he did but he didnt try gradient boosting. it wouldnt have made much of a difference tho, probably.
by talkingpractice
Tue Aug 05, 2014 6:13 pm
Forum: APBRmetrics
Topic: Estimating Replacement Level: Sabermetrics perspective
Replies: 33
Views: 25527

Re: Estimating Replacement Level: Sabermetrics perspective

using a multi-year reference group comprised of the low mp players What would be a "low mp player" in your approach? What kind of threshold is used to determine that? Also, if you apply such a threshold to the group of reference players for those minimum contracts, how many of those would fall into...
by talkingpractice
Tue Aug 05, 2014 5:55 pm
Forum: APBRmetrics
Topic: Estimating Replacement Level: Sabermetrics perspective
Replies: 33
Views: 25527

Re: Estimating Replacement Level: Sabermetrics perspective

How many total possessions did those replacement players play? It might be a good idea to use a lower lamdba for the replacement player column, as the coefficient might be dragged "too far" towards zero when there's not a lot of data it was actually like over 200k possessions or something. i was re...
by talkingpractice
Mon Aug 04, 2014 3:46 pm
Forum: APBRmetrics
Topic: Estimating Replacement Level: Sabermetrics perspective
Replies: 33
Views: 25527

Re: Estimating Replacement Level: Sabermetrics perspective

well... the results for the RAPM part of this exercise only add more clarity to the "replacement level is higher than we thought" argument. first, the methodology: 1, we used 6 years of data (2009-2014), and made a reference group out of all of the playerseasons of minimum-salary players. 2, we used...
by talkingpractice
Thu Jul 31, 2014 10:44 pm
Forum: APBRmetrics
Topic: Estimating Replacement Level: Sabermetrics perspective
Replies: 33
Views: 25527

Re: Estimating Replacement Level: Sabermetrics perspective

i'll have the rapm version of this for us by Monday at the latest.
by talkingpractice
Wed Jul 30, 2014 11:53 pm
Forum: APBRmetrics
Topic: Estimating Replacement Level: Sabermetrics perspective
Replies: 33
Views: 25527

Re: Estimating Replacement Level: Sabermetrics perspective

aaaaaaaaaaaahhhhhhhh, of course. im a bozo.
by talkingpractice
Wed Jul 30, 2014 6:51 pm
Forum: APBRmetrics
Topic: Estimating Replacement Level: Sabermetrics perspective
Replies: 33
Views: 25527

Re: Estimating Replacement Level: Sabermetrics perspective

what is meant by the Preseason row?

one of my guys already confirmed that there are way less naming discrepancies between these lists and our pbp than i guessed there would be originally, so ill have the rapm version of this back much quicker than i said earlier.