Search found 912 matches

by EvanZ
Wed Jun 01, 2011 10:55 am
Forum: APBRmetrics
Topic: APBR Retrodiction Challenge for Summer 2011
Replies: 74
Views: 29115

Re: APBR Retrodiction Challenge for Summer 2011

It's up to each metric to decide how to handle rookies. You know, whatever gives you the best predictive power.

Obviously, the only restriction is that you can't use any 2010-11 data to fit your model.
by EvanZ
Tue May 31, 2011 10:47 pm
Forum: APBRmetrics
Topic: APBR Retrodiction Challenge for Summer 2011
Replies: 74
Views: 29115

Re: APBR Retrodiction Challenge for Summer 2011

Sure, use playoff data. Anything known before opening night is fair game.
by EvanZ
Tue May 31, 2011 9:46 pm
Forum: APBRmetrics
Topic: Looking for 2011 playoffs lineup data
Replies: 16
Views: 8817

Re: Looking for 2011 playoffs lineup data

I am going to need to figure out how to take the play by play data and convert it so I can do this on my own. Do you know Ruby? I can give you my code. As long as you follow the CC license, I'm happy to do that. I do not know Ruby. Is it something I can pick up with some work? I greatly appreciate ...
by EvanZ
Tue May 31, 2011 9:12 pm
Forum: APBRmetrics
Topic: Looking for 2011 playoffs lineup data
Replies: 16
Views: 8817

Re: Looking for 2011 playoffs lineup data

Nima wrote: I am going to need to figure out how to take the play by play data and convert it so I can do this on my own.
Do you know Ruby? I can give you my code. As long as you follow the CC license, I'm happy to do that.
by EvanZ
Tue May 31, 2011 9:09 pm
Forum: APBRmetrics
Topic: Looking for 2011 playoffs lineup data
Replies: 16
Views: 8817

Re: Looking for 2011 playoffs lineup data

Yes, it's what Crow said. Those other players are the opponents that Dallas faced in those lineups. It actually gives you the ability to determine opponent quality. It doesn't include playoff data. I'm not sure you would even benefit from that, since it's a) small sample size and b) all against one ...
by EvanZ
Tue May 31, 2011 6:59 pm
Forum: APBRmetrics
Topic: Looking for 2011 playoffs lineup data
Replies: 16
Views: 8817

Re: Looking for 2011 playoffs lineup data

With Dallas' lineups, the 5 I would look at would certainly be the 5 that have played the most possessions in the playoffs. 1. Kidd, Stevenson, Marion, Nowitzki, Chandler 2. Barea, Terry, Stojakovic, Nowitzki, Haywood 3. Kidd, Terry, Marion, Nowitzki, Chandler 4. Kidd, Terry, Stojakovic, Marion, Ch...
by EvanZ
Tue May 31, 2011 4:59 pm
Forum: APBRmetrics
Topic: Looking for 2011 playoffs lineup data
Replies: 16
Views: 8817

Re: Looking for 2011 playoffs lineup data

It can be done with some effort, maybe not by tonight, though.

What lineups are you interested in? It's a manual process right now, so it would take about 5 minutes for me to do each one. I can probably do 10 or so for you.
by EvanZ
Tue May 31, 2011 4:56 pm
Forum: APBRmetrics
Topic: APBR Retrodiction Challenge for Summer 2011
Replies: 74
Views: 29115

Re: APBR Retrodiction Challenge for Summer 2011

... do a retrodiction of 2010-11 win totals using 2009 player ratings and actual minutes played in 2010-11... ... Obviously, the only criterion is that you can't use any 2010 data besides minutes. . Clarification needed. To me, 2010 refers to the 2009-10 season. That's what I'd always use to predic...
by EvanZ
Tue May 31, 2011 4:54 pm
Forum: APBRmetrics
Topic: APBR Retrodiction Challenge for Summer 2011
Replies: 74
Views: 29115

Re: APBR Retrodiction Challenge for Summer 2011

I don't really like team efficiency differential. One might predict a good differential for Boston because his system says Glen Davis is great. Per possession error is more useful I think. I'll go along with team efficiency differential though http://www.basketballvalue.com/publicdata/matchups20102...
by EvanZ
Tue May 31, 2011 4:32 pm
Forum: APBRmetrics
Topic: APBR Retrodiction Challenge for Summer 2011
Replies: 74
Views: 29115

Re: APBR Retrodiction Study for Summer 2011

Also, think of this as an iterative challenge. You can submit as many results as you like. The goal is to get better and better, if possible... Or for someone like me, maybe just to see how far behind I am. :lol: Also, we should use a "master" possession file so that everyone is working with exactly...
by EvanZ
Tue May 31, 2011 4:31 pm
Forum: APBRmetrics
Topic: APBR Retrodiction Challenge for Summer 2011
Replies: 74
Views: 29115

Re: APBR Retrodiction Study for Summer 2011

Yeah, I was thinking squared error in differential, too.

Regular season only.
by EvanZ
Tue May 31, 2011 4:17 pm
Forum: APBRmetrics
Topic: APBR Retrodiction Challenge for Summer 2011
Replies: 74
Views: 29115

APBR Retrodiction Challenge for Summer 2011

Now that the regular season is over, I want to propose that each of us with rating systems do a retrodiction of 2010-11 win totals using 2009 player ratings and actual minutes played in 2010-11. Actually, you could use any prior-year rating system you like. The goal would be to see who comes up with...
by EvanZ
Mon May 30, 2011 2:51 pm
Forum: APBRmetrics
Topic: Calculated Positions
Replies: 6
Views: 3860

Re: Calculated Positions

DSMok1 wrote:
Evan: is this for offense or defense or just whatever? Defense can be more useful to know, but may be less cut-and-dried. Good defenders will cross-match onto whomever is the best opposing player.
No, but that would be extremely useful. Somebody would have to track that.
by EvanZ
Mon May 30, 2011 1:51 pm
Forum: APBRmetrics
Topic: Calculated Positions
Replies: 6
Views: 3860

Re: Calculated Positions

Aaron B. made initial estimates of each player's position, and then for each unit, the players are ranked according to that nominal position. What I'm doing is simply iterating on that ranking. I can give one example where it shows the improvement. Stephen Curry is a 1.5 according to the initial est...
by EvanZ
Mon May 30, 2011 1:14 pm
Forum: APBRmetrics
Topic: Calculated Positions
Replies: 6
Views: 3860

Calculated Positions

I remember volunteering to calculate positions from Aaron B's matchup files a while back. Here they are for 2010-11 (csv): http://tinyurl.com/3n5qkmb The column "CPOS" is the calculated position based on the following formula: calculated_position=(player.pos1+2.0*player.pos2+3.0*player.pos3+4.0*play...