About

The Arbitrarian is the work of David Sparks, a graduate student in Political Science at Duke University. I’m interested in politics, statistics, information design, sports, and everything else. Feel free to visit my personal homepage at Duke, or see a list of other work I’ve produced.

Advertisements

16 responses to “About

  1. Although I don’t really understand half of what I have seen from you , it is quickly evident that you are a bright guy. I coach a girls varsity basketball team and I was purusing the web to try and find a decent stat comparison spreadsheet to fairly evaluate and grade my players overall contribution to the team. I stumbled upon your site. Good stuff.

  2. David, I’m very interested in the People’s Statistic Project. I fool around with stats myself, but not the extent that you and other APBR folks do so. However, when I first looked at the PS results, it occurred to me that it looked more like a high-concept popularity contest than an actual “contributes to wins” stat. From a marketing point of view, I think that might actually have some value. You mentioned in a forum that many of the surveyed seem to be trying to tweak the results to make MJ the #1 all-time. I think if you included dunks in the survey that might put him over the top, and yield a weighting system that could predict a players “star-power” in the modern age. Chris Paul is becoming very popular, but can you imagine if he dunked on guys? ๐Ÿ˜‰

  3. Just thought you would like to know I have posted your interview. It can be found at http://3shadesofblue.blogspot.com/2008/05/arbitrary-interview-with-arbitrarian.html

  4. David,

    The contribution you just had published on Hardwood Paroxysm … is OUTSTANDING.

    Bravo!

    A Tour de Force.

  5. Secondly.

    Given the respective weights you’ve mentioned, thus far, am I reading your work correctly … if I assert from this initial material that the two measures which receive the most weight in your metric are:

    * or 1.0192830

    * ‘total rebounds’ [or/1.0192830 + dr/ 0.5173558]

    [considering the number of ‘rebounds’ there are in a game, relative to the number of ‘other events’ which happen]

  6. khandor: Interesting idea, to look at the stats from which the most points were gained or lost. Here’s the breakdown, summing statistics since 1979-80:

    pts: 6384067
    fgx: -2858370
    ftx: -240559.2
    as: 1631794
    or: 839599.7
    dr: 953948.7
    st: 828238.8
    bk: 312690.8
    to: -1401794
    pf: -320387.7

    So, the biggest source has been scoring (somewhat obviously), and the biggest drain has been missed fgas. It appears that defensive rebounds have contributed more than offensive boards, due to the much greater number of drs than ors.

  7. David,

    Had an interesting exchange with Kevin Broom a while back, re: the relationship between ‘rebounds garnered in a specific game’ and ‘winning that game’ based on ‘regression analyses’ of traditional boxscore information and his adherence to the ‘4 Factors’ theories of Dean Oliver.

    Mentioned to him that some years ago (actually in the mid ’90’s), as best I could recall, the research studies I read indicated that the most SIGNIFICANT relationship existed between ‘rebounds garnered in a specific game’ and ‘winning that specific game’ when using regression analysis of traditional boxscore stats (not including actual Pts Scored, of course, as you’ve identified above).

    Kevin indicated that, according to the work he’d seen, my perception was inconsistent with the information he had at-hand (i.e. that of Dean Oliver and others).

    Based on your specific metric and your use of regression analysis would you be able to corroberate my perception, in this regard, or not?

  8. David,

    What happened to your article, “Anything is Possible” in your ‘Top Posts’ section?

    It was there on last week … but now it’s disappeared.

  9. Khandor: I’m not sure why it dropped off the Top Posts lists, but you can find it here: https://arbitrarian.wordpress.com/2008/06/18/anythings-possible/

    Sorry it’s taking me so long to follow up on your question about rebounding… between writing this week’s HP post and Real Life, I’m pretty busy.

  10. Hi, David.

    Thanks a bunch.

    No problem, from my end … Me, too, in fact. ๐Ÿ™‚

    Too much to handle, all at once.

    ASAP … will be just fine. ๐Ÿ™‚

  11. Solid job, once again, with your latest contribution over at HP.

    When you get a chance, please remember my question in this section. ๐Ÿ™‚

    Thanks

  12. Hi, David.

    Still no answer worthcoming?

    [just a friendly reminder ๐Ÿ™‚ ]

  13. Hi, David.

    Still no answer forthcoming? ๐Ÿ˜ฆ

  14. Hi, David.

    Is there going to be an answer provided for the question I asked?

    Hopefully there is. ๐Ÿ™‚

  15. Hi, David.

    I’m still waiting for an answer to my specific question. ๐Ÿ™‚

  16. Hi there,

    I couldn’t find a “contact us” page, so I am writing to you this massage here:

    I run the service R-bloggers:
    http://www.r-bloggers.com/about/
    An aggregator of R related articles, from blogs.

    And wanted to encourage you to join R-bloggers.com at:
    http://www.r-bloggers.com/add-your-blog/

    The idea behind the project is to share readers in order to gain readers: R-bloggers already has over 800 RSS subscribers (that are growing everyday).

    I built it in order to find all the R bloggers out there. So far I found over 45 bloggers, which also agreed to add there feed (and some to give a link back and post about it).

    And would love it if you might agree to join as well.

    Feel free to erase this comment if it clutters the blog too much.

    All the best,
    Tal

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s