Loading ...
Sorry, an error occurred while loading the content.
Skip to search.

Re: [APBR_analysis] Re: New file uploaded to APBR_analysis

Expand Messages
  • Michael K. Tamada
    On Mon, 24 Dec 2001, HoopStudies wrote: [...] ... Mike G s critique I think makes a number of worthwhile points. One of the most important ones is the
    Message 1 of 77 , Jan 3, 2002
      On Mon, 24 Dec 2001, HoopStudies wrote:


      > This is the article by Dave Berri, an Econ prof at Cal St. San
      > Bernadino (was at Coe College when wrote this). It is an
      > econometrics look at who created the most wins for their team in
      > 1998. It is a formal method that is worthwhile spending some time
      > with. I can explain some of the techniques, though the Cobb-Douglas
      > form mentioned in here was something new to me.
      > Most people will jump to the results, which showed that, uh, Dennis
      > Rodman produced the most individual wins in 1998.

      Mike G's critique I think makes a number of worthwhile points. One of the
      most important ones is the linkage, or lack thereof, between and
      individual player's performance and his team's performance.

      If I'm reading the article correctly, Prof. Berri used data on teams to
      estimate the value of items such as points, Points Per Shot (PPS),
      rebounds, etc. As far as I can tell, he then looked at individual
      players' accumulation of points, PPS, rebounds, etc. to rate their
      performances (plus making allowances for game pace, etc.). But he used
      the team weights to evaluate individual players' statistics. This is
      almost certainly not a good procedure. Here's an example:

      From a team perspective, PPS (or FG%, or other such efficiency measures)
      are going to be very important, more important than FGM or points scored
      (since those measures are contaminated by game pace). Which is the better
      offensive team, one which scores 110 points per game while shooting 40%,
      or one which scores 105 points per game while shooting 50%? Prof. Berri's
      research shows, probably correctly, that PPS is more important when
      evaluating teams' offenses.

      But, as Mike G pointed out, it is almost surely a mistake to apply these
      team weights to individual players. Points scored and PPS (or FG%) being
      the prime example, one which several of us have mentioned before. A 55%
      FG% player such as Bo Outlaw or the elderly Artis Gilmore may sound great,
      but if they're only scoring 8 points a game, they aren't really helping
      the offense very much. Conversely, an Alan Iverson can be helping his
      team, even with his wretched 42% FG%. (Although he still did not deserve
      that MVP award he won last year.)

      What is missing from Prof Berri's article is a true model of how
      individual players' contributions lead to overall team success (or lack of
      success). This is IMO the Holy Grail of sports statistics research, most
      especially basketball research. It is unfortunately exceedingly
      difficult. Even without fancy statistics or models though, there are a
      couple of obvious features that such a model must have:

      1. Despite Prof Berri's belittling of assists, it is intuitively obvious
      that a team has to have at least one skilled ballhandler (or 2 or 3
      semi-skilled ones) on the floor. Even the Don Nelsons of the NBA who like
      to go to tactical extremes don't put 3 centers and two power forwards on
      the floor at one time (something which probably makes statistical sense in
      models such as Prof Berri's, because those big guys get more rebounds and
      usually shoot a higher percentage than the little players).

      [Side note: Peter May a couple of weeks ago wrote for espn.com an
      encomium to Antoine Walker, who despite his lousy FG% and predilection for
      jacking up 3-pointers clearly is almost as important a part of the
      Celtics' success this year as Paul Pierce. It occurs to me that while
      players such as John Johnson, Paul Pressey, and Scottie Pippen pioneered
      the concept of the ball-handling, playmaking, "point forward", Antoine
      Walker may be the first "point power forward" in NBA history. Bird of
      course was a playmaking machine, but I consider him a small forward rather
      than a power forward. I suppose Chris Webber should get some mention
      here too, although I think of him as more of a post player than Walker.]

      2. As we discussed on this list several weeks ago, a team with five 55%
      FG% shooters might look great on paper, but if they're all a bunch of Todd
      MacCullochs, they're going to be an awful offensive team. Conversely a
      team with 5 Alan Iversons will probably also be a bad offensive team,
      although if they learn to share the ball and restrict themselves to good
      shots, they have the potential to be a very good one.

      But having one Todd MacCulloch, or one Alan Iverson, can be a good thing,
      as it was for the 76ers last year. It depends on who are the other
      players in your lineup.


      P.S. The Cobb-Douglas functional form has an interesting history. Paul
      Douglas was an economist at the University of Chicago, who went on to
      enter politics and become a US senator, and prior to that a Chicago
      alderman (city councilman), most remarkably an honest Chicago alderman.

      When he was still an econ professor, he was trying to model production
      functions, the relationship between output (manufacturing output in his
      case) and factor inputs (capital and labor). He couldn't think of a
      mathematical function that would model them adequately, so he went to a
      math professor, Charles Cobb who suggested the functional form now known
      as a Cobb-Douglas form.

      In its simplest form, if "Q" is the quantity of output produced in a given
      period of time, "L" the amount of labor used, and "K" the amount of
      physical capital used, then the Cobb-Douglas form assumes that

      Q = b * L^a * K^(1-a)

      It's usually more convenient to take logarithms of these variables, so

      ln(Q) = ln(b) + a*ln(L) + (1-a)*ln(K)

      which is a nice simple linear equation which has a number of nice
      mathematical characteristics. Probably too nice, as real world production
      functions are more complex than what is contained in these equations.
      For example it assume that there are constant returns to scale, and thus
      no economies of scale.
    • APBR_analysis@yahoogroups.com
      Hello, This email message is a notification to let you know that a file has been uploaded to the Files area of the APBR_analysis group. File : /Warriors
      Message 77 of 77 , May 5, 2005

        This email message is a notification to let you know that
        a file has been uploaded to the Files area of the APBR_analysis

        File : /Warriors Stats.pdf
        Uploaded by : skauffman <skauffman@...>
        Description : Analysis of the Golden State Warriors 2004-05 Season

        You can access this file at the URL:

        To learn more about file sharing for your group, please visit:


        skauffman <skauffman@...>
      Your message has been successfully submitted and would be delivered to recipients shortly.