Search found 194 matches

by talkingpractice
Mon Apr 21, 2014 10:09 pm
Forum: APBRmetrics
Topic: MIP ie MIS Award
Replies: 7
Views: 3900

MIP ie MIS Award

So, I think the MIP discussions are superannoying, since in almost all cases, the increased observed performance for the player is much more about context than about improvement. Hence, the MIS Award (Most Improved Situation) should go to lots of the people I've been seeing on most MIP lists. To try...
by talkingpractice
Mon Apr 21, 2014 8:39 pm
Forum: APBRmetrics
Topic: 2013-14 Rookies
Replies: 54
Views: 23158

Re: 2013-14 Rookies

If anyone is curious, the best rooks by IPV were -> ROOK Michael Carter-Williams = 0.642079917 ROOK Victor Oladipo = -0.210964858 XROOK Pero Antic = -0.30719532 ROOK Kelly Olynyk = -0.324575169 ROOK Matthew Dellavedova = -0.341016326 Some other notable traditional rooks -> ROOK Jeff Withey = -0.567...
by talkingpractice
Tue Apr 08, 2014 12:34 am
Forum: APBRmetrics
Topic: Well, well, well...
Replies: 42
Views: 21983

Re: Well, well, well...

Obviously I think this whole thing is awesome. J.E., if this is xRAPM why is there absolutely no mention at all of the boxscore component in the article? According to you, its 65% of the metric but its not mentioned in the article. If someone read the article, they would assume its some kind of NPI ...
by talkingpractice
Wed Mar 19, 2014 3:33 pm
Forum: APBRmetrics
Topic: Ibaka
Replies: 16
Views: 4746

Re: Ibaka

He's #31 right now in IPV (boxscore/demographic informed RAPM) at +2.9, right below Wall, and right above Kawhi.

Though I'm not a big fan of using these as rankings per se (same with RAPM in general) due to how context-specific this stuff is.

JE's good blocks vs bad blocks comment is mad cool.
by talkingpractice
Tue Mar 18, 2014 9:40 pm
Forum: APBRmetrics
Topic: Ibaka
Replies: 16
Views: 4746

Re: Ibaka

It confuses me a bit too, in terms of the eye test. Elite rim protection, elite mid range shooting, great spacing from/for a big, and he's a way improved pnr defender now. But as JE/Crow said, OKC just doesn't seem to play any better with him than they do without him. His on/off (and hence RAPM valu...
by talkingpractice
Sun Mar 16, 2014 4:24 pm
Forum: APBRmetrics
Topic: 2014 Draft Projection Models
Replies: 102
Views: 76953

Re: 2014 Draft Projection Models

I think that poor Otto is (thus far) a casualty of really bad luck and pretty bad coaching, namely: - 2 guys having much better years than expected (Ariza, Webster), - his team spending most of the year fighting for 3rd while fearing 9th, - WAS not (yet) having a D League team all to themselves, and...
by talkingpractice
Mon Mar 03, 2014 11:47 pm
Forum: APBRmetrics
Topic: Statistical +/-
Replies: 10
Views: 8555

Re: Statistical +/-

I have various comments here, and hopefully some will be helpful -> - So our FORPM model (we've discussed it a bit but not with tons of detail) uses random forest regressions and gradient boosting, for many of the reasons that Ilardi mentioned. Our findings match a lot of what I'm reading here, name...
by talkingpractice
Mon Feb 24, 2014 6:15 pm
Forum: APBRmetrics
Topic: The 'effect of being up X'
Replies: 36
Views: 34683

Re: The 'effect of being up X'

If it's at all helpful or interesting, we've (ofc) tried out various ways of operationalizing a garbage time cutoff of some sort in our RAPM/similar models, and always find that any benefit (to out of sample predictive power) is slim to none. This is probably due to the shrinking sample sizes from d...
by talkingpractice
Wed Feb 19, 2014 5:23 pm
Forum: APBRmetrics
Topic: Prior-Informed RAPM (Updated 2-3 Times/Wk)
Replies: 40
Views: 9766

Re: Prior-Informed RAPM (Updated 2-3 Times/Wk)

JE, can you explain what you mean by 'effect of leading' adjustment?
by talkingpractice
Fri Feb 14, 2014 3:57 pm
Forum: APBRmetrics
Topic: Prior-Informed RAPM (Updated 2-3 Times/Wk)
Replies: 40
Views: 9766

Re: Prior-Informed RAPM (Updated 2-3 Times/Wk)

Looking at talkingpractice’s RAPM and JE’s from last week I see some cases of divergent player estimates. I know that the methods are different (and the # games covered might be too) so they are not going to be identical but I think a comparison is still worthwhile. A few extreme differences: On TP...
by talkingpractice
Sat Feb 08, 2014 11:48 pm
Forum: APBRmetrics
Topic: Rough early estimate of chances to win this season's tiltle
Replies: 33
Views: 11230

Re: Rough early estimate of chances to win this season's til

Here's the current odds on Betfair converted to P(win), if anyone is interested.

MIA = 27%
IND = 24%
OKC = 19%
SAS = 8%
LAC = 7%
GSW = 4%
POR = 4%
HOU = 4%

no one else better than 75:1.
by talkingpractice
Sat Feb 08, 2014 3:33 pm
Forum: APBRmetrics
Topic: Kirk Goldsberry Article & Sloan Paper - Databall
Replies: 25
Views: 22238

Re: Kirk Goldsberry Article & Sloan Paper - Databall

Everyone is being way too rough on EPV imo. It takes time to go from the original idea to the final version (of anything like this).

I do think that they pulled a McDowells (Coming to America reference) in regards to their naming of the metric ;)
by talkingpractice
Sat Feb 08, 2014 3:26 pm
Forum: APBRmetrics
Topic: Prior-Informed RAPM (Updated 2-3 Times/Wk)
Replies: 40
Views: 9766

Re: Prior-Informed RAPM (Updated 2-3 Times/Wk)

Excuse me for piggybacking You're far from a piggyback. Happy to have you in any thread of ours. I'll probably update weekly - these numbers don't really change much from day to day. Ftr, we're only going to update once/week or so too now. Doing it 3 times/week isn't really needed, for the same rea...
by talkingpractice
Mon Feb 03, 2014 11:23 pm
Forum: APBRmetrics
Topic: Prior-Informed RAPM (Updated 2-3 Times/Wk)
Replies: 40
Views: 9766

Re: Prior-Informed RAPM (Updated 2-3 Times/Wk)

Quick comment on some of the things mentioned here. These prior-informed (by previous year) RAPMs are biased (ofc) to that selection of prior. The priors here are set very methodically (for purposes of this list we publish). So for most of the 2014 rookies to be near -2 is not surprising, as most ha...
by talkingpractice
Wed Jan 22, 2014 3:43 pm
Forum: APBRmetrics
Topic: Early 2013-14 stat-based observations
Replies: 68
Views: 14086

Re: Early 2013-14 stat-based observations

I looked up random forests and saw that is considered a very helpful approach overall. I did see this caveat at the wikipedia page: "This method of determining variable importance has some drawbacks. For data including categorical variables with different number of levels, random forests are biased...