Re: New file uploaded to APBR_analysis
- My observations/questions-
1. He does seem to criticize linear weights, but I think he meant to
criticize specific formulas such as Heeren's and Bellotti's for their
lack of empirical evidence. Afterwards, he does propose a model of
2. As far as I can tell, he then uses certain statistical tests to
try to show how this linear model fits the data better than the Cobb-
Douglas model. Does anyone know of other similar tests? I'm guessing
that there are some other formulas that test a proposed model like
3. Assuming his statistical tests are sound, he seems to demonstrate
that his linear model is better than the Cobb-Douglas framework. In
page 3, he seems to associate the latter with the idea that the value
of a given stat depends on a team's other statistics. Did he intend
to debunk this idea or just the Cobb-Douglas model that was
associated with it? It seems to me that Berri's linear concept
doesn't seem compatible with the concept of player performances being
interdependent with team performance. For example, wouldn't Iverson's
misses have more negative value if he were on a team with lousy
4. I noticed in the team tempo equation that FG attempts are weighted
more heavily than FT attempts by a ratio close to that found in Dr.
Oliver's team possession formula. Is this notable?
5. Berri's solution to the apparent statistical bias towards big
players doesn't seem to fit well with the rest of his theory. It's
hard for me to explain my unease since I don't have rigorous training
in statistics, but I feel that his framework seems to render such
adjustments unnecessary. For all I know, his model could really be
saying that big players are a lot more valuable than smaller players
(He writes about big players getting more rebounds and the smaller
guards getting more turnovers in his section on positional
adjustments). If so, maybe this suggests a deficiency in the
statistics we keep, a flaw in his model, or that big players really
are much better.
6. I don't know if this is one of the formula's intended functions,
but what does the formula say about the performance of the San
Antonio Spurs in the season when David Robinson had that long absence
due to injury?
7. Do the parentheses around some numbers mean that they're negative?
I'm having trouble with the figures in Tables 6-8.
8. The paper deals with "wins producers." What might a study on "loss
producers," given this framework, yield?
--- In APBR_analysis@y..., "HoopStudies" <deano@r...> wrote:
> Thanks for the comments. Keep them coming.
> I generally passed on the same comments to Dave before I got the
> article posted. I do think it is important that we better
> his work and he better understand ours. He, for instance, knew of
> Bill James but no details of his work. Econometric guys have their
> framework for the world and, in this case, he tried to make
> basketball fit that framework. I certainly don't think the
> fits the individual interactions of the game, though, his analysis
> shows that team numbers can be replicated with his model pretty
> well. I am struggling a bit with the communication to get him to
> understand why I don't think the method works -- other than the
> conclusions being "obviously untenable" (I think I said that the
> results don't pass the laugh test). I do not think that his team
> analysis can be simply broken down into individuals, but I think my
> reason right now is that, uh, the results aren't good. I need a
> better reason than that and I'm working to figure out exactly where
> his logic is flawed. (Maybe asking him to explain the Iverson
> of a couple messages ago is a start.)
> Anyway, couple things
> 1. I will pass on your comments to Dave as soon as we get a good
> of them.
> 2. I think it is important to interact with the academic world
> this kind of work. He posed some good objectives in his intro --
> looking at "the extent of racial discrimination" and "the impact
> alternative institutional arrangements have on worker compensation
> professional basketball".
> 3. He bashes linear weights methods like Tendex, etc., though I
> believe his method essentially is that, but generating a number
> means something -- wins. He effectively generates weights for each
> statistic based on the correlation to wins. A lot of people here
> like to rely on linear weights -- anyone want to specifically
> on how he did it, whether it has any relevance to other methods,
> whether his stuff could be tweaked?
> One of the things I generally do not like actually is that he does
> evaluate players against the average of their position. I like
> absolute measures that perhaps show that big men dominated in the
> 70's, point guards in '80's, and swingmen-size players since. That
> tells us something about how the game has changed. Normalizing to
> positional averages hides it (and, I think for this case, hides a
> flaw in his method).
> Dean Oliver
> Journal of Basketball Studies
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: