Search found 294 matches

by xkonk
Fri Jan 01, 2016 4:39 am
Forum: APBRmetrics
Topic: Trivia
Replies: 228
Views: 285219

Re: Trivia

I know you listed Boston as best in the East because of their SRS, but they have to have about the weirdest distribution of point differentials I've seen. They hardly play close games. Which is more believable: their SRS, or their fivethirtyeight ELO (http://projects.fivethirtyeight.com/2016-nba-pic...
by xkonk
Tue Dec 01, 2015 12:59 am
Forum: APBRmetrics
Topic: Team contact information
Replies: 4
Views: 4060

Re: Team contact information

Some teams put at least some info on their websites. I was able to look up a couple of people when I applied with the Hornets a bit ago, for example: http://www.nba.com/hornets/staff-directory . Presumably David Kaplan and Nick Repole are part of the analytics team, although I suppose they could be ...
by xkonk
Wed Nov 25, 2015 3:27 am
Forum: APBRmetrics
Topic: calculating the diminishing returns of team quality
Replies: 8
Views: 5730

Re: calculating the diminishing returns of team quality

The values on Evan's chart from -3 to 0 all outperformed the slope line. The values just beyond -3 all underperformed. I think the threshold for winning a decent % of roads is a relevant and key factor as I suggested above. That's just the graph for one season. I grabbed a much larger data set (the...
by xkonk
Tue Nov 24, 2015 1:01 am
Forum: APBRmetrics
Topic: calculating the diminishing returns of team quality
Replies: 8
Views: 5730

Re: calculating the diminishing returns of team quality

I've always thought the limit on games that could be won (the max of 82) should be taken into account, as you say, but I think that practically speaking it doesn't matter. Evan Z has a post (that I bring up just because I remembered it; he isn't the originator) showing the correspondence between win...
by xkonk
Tue Nov 24, 2015 12:34 am
Forum: APBRmetrics
Topic: 2015-16 Team win projections
Replies: 322
Views: 263004

Re: 2015-16 Team win projections

Agreeing with ca1294, my understanding has always been that squared errors are preferred because that produces a function that can be differentiated, which is important for the calculus that underlies statistics (hence variance and standard deviation but also higher order moments like skewness or ke...
by xkonk
Mon Nov 23, 2015 1:30 am
Forum: APBRmetrics
Topic: Win Probably Calculator ??
Replies: 1
Views: 2121

Re: Win Probably Calculator ??

I'm a little confused by how your question. Are you trying to predict WPA from box score stats? What is in the database you found exactly? At any rate, here's one calculator, but I've seen others as well: http://stats.inpredictable.com/nba/wpCalc.php . Unless your son plays in the NBA though, I doub...
by xkonk
Fri Nov 13, 2015 11:43 am
Forum: APBRmetrics
Topic: RAPM adjusted for luck?
Replies: 11
Views: 9725

Re: RAPM adjusted for luck?

Anywho - Out-of-sample testing for individual stats like this would be huge for RAPM I think. RAPM, or at least the one that Jerry has published, uses cross validation/out-of-sample testing to adjust how much regression to the mean occurs. Is this not enough of an adjustment for luck? Or are you al...
by xkonk
Fri Oct 30, 2015 11:20 pm
Forum: APBRmetrics
Topic: Can players learn from BPM?
Replies: 16
Views: 10572

Re: Can players learn from BPM?

Targeting assists would likely not be so problematic, as there is no direct negative trade off potential. As Mike noted above, what about turnovers? Also, it would certainly be more difficult for some players to get assists because they would have to get the ball in their hands in the first place. ...
by xkonk
Sat Oct 24, 2015 2:58 pm
Forum: APBRmetrics
Topic: Advanced Free Throw Analytics
Replies: 5
Views: 4992

Re: Advanced Free Throw Analytics

Instead of jumping around to miss% to try and get to a ratio, just use the odds ratio. That's what it's for. And it makes your multiplicative statements ('twice as likely') actually meaningful. Sessions has the best odds ratio in increasing from first to second free throw; the odds are ~2.8 times as...
by xkonk
Tue Aug 18, 2015 12:02 am
Forum: APBRmetrics
Topic: Guides to Creating RAPM
Replies: 27
Views: 26393

Re: Guides to Creating RAPM

Here's what I used: X = 39192 x 479 lineup matrix y = 39192 x 1 results vector (only includes stints where both home and away teams got a possession, value is difference between home and away points per 100 possessions) weights = 39192 x 1 vector of observation weights (sum of home and away possess...
by xkonk
Thu Aug 13, 2015 11:52 pm
Forum: APBRmetrics
Topic: Guides to Creating RAPM
Replies: 27
Views: 26393

Re: Guides to Creating RAPM

It's been a while since I've looked at RAPM-type data... can you spell out what the non-x variables are? In particular, how did you make y and the weights? I assumed y is point differential per 100 possessions and weight is the number of possessions for a particular match-up, but then I don't think ...
by xkonk
Fri Jul 17, 2015 12:37 am
Forum: APBRmetrics
Topic: Diversity vs. Quality
Replies: 11
Views: 5888

Re: Diversity vs. Quality

In terms of front offices, the paper seems relevant. In regards to teams themselves, I'm not so sure. The key point is that in the limit , random beats out individual quality. If we make the leap from their model of decision making to playing basketball, we still have to think that five players is a...
by xkonk
Sat Jul 11, 2015 12:56 am
Forum: APBRmetrics
Topic: Worst a Player Can Be
Replies: 27
Views: 17436

Re: Worst a Player Can Be

What's the difference between those two graphs? Just the scale of the y axis? I don't see the exponential on the top graph, and it looks like it pretty much follows the gamma in the bottom, but none of the fits you have on there seem to describe the actual NBA data very well. Looks like you need som...
by xkonk
Fri Jul 10, 2015 1:02 am
Forum: APBRmetrics
Topic: Worst a Player Can Be
Replies: 27
Views: 17436

Re: Worst a Player Can Be

If you think it's a gamma, can't you pull some random samples and compare them to actual NBA ratings to see what looks close? Then you'll have the distribution and will know the asymptote. Trying to figure out how to do this. Right now I'm investigating what kind of curves fit well the right tail o...
by xkonk
Thu Jul 09, 2015 3:19 pm
Forum: APBRmetrics
Topic: Hot Hand
Replies: 20
Views: 17563

Re: Hot Hand

The same authors have a paper (linked in http://andrewgelman.com/2015/07/09/hey-guess-what-there-really-is-a-hot-hand/ ) showing/explaining how certain methods of calculating a hot hand underestimate the value. For example, the classic Gilovich study that found a statistically non-significant hot ha...