From: Dean Oliver [mailto:deano@...
Sent: Wednesday, July 30, 2003 11:43 AM
--- In APBR_analysis@yahoogroups.com
, "Kevin Pelton" <kpelton08@h...>
>> saved the best for last here, DeanO. For me, it was impossible to
>> read _Moneyball_ without thinking of its NBA implications (only in
>Quality of competition has already rung true, though. James' games in
>Orlando and Boston weren't as good as some had hoped. His athleticism
>didn't shine like it did in HS, but it was clearly still good. His
>fundamentals are what scouts ended up talking about there, which was a
>big change from before. Quantify that!
This repeats some things we've discussed before, but they're probably
worth repeating: despite baseball's overall greater amount of
quantitative stats and ease of being quantitatively analyzed (compared
to basketball), sabrmetrics techniques can probably add more to the
study of baseball than hoop-o-metric techniques can add to the study
of basketball. Because there's some pretty good sabrmetric evidence that
a lot of baseball truisms, myths, received wisdom, and ways of doing
business are wrong, and can be improved. In addition, when it comes to
drafting young players, baseball is in a much much tougher situation
than basketball, and the draft is much more of a crapshoot in baseball.
Mike Piazza was what, a 63rd round draft choice? Conversely baseball
is filled with David Clydes and other high pick bonus babies who were
total busts in the majors.
Drafting basketball players is much easier. How much statistical
digging did Milwaukee need to do when Lew Alcindor was a senior and
they won the coin flip? Except for LaRue Martin, a number one NBA
draft choice is pretty much guaranteed to become an NBA starter quality
player, and usually an all-star (with occasional exceptions such as
Olowakandi, Joe Smith, etc.).
That's not to say that drafting NBA players is easy, just easier than
baseball. Sam Bowie or Michael Jordan? The choice was non-trivial
when Portland faced it. But, as DeanO says, quantify that. I think
we're a long way from being able to provide big help the Portlands
and other teams facing tough draft decisions. I'm sure that DeanO
and other consultants provide some help. But not big help.
All that said, I think one area worth exploring (maybe others have
already explored this and found it hopelessly difficult) is working
out translation factors for NCAA players into NBA players, just as
sabrmetricians have worked out translations for minor leaguers into
major leaguers (and now, even Japanese leaguers into major leaguers).
Other people on this list have mentioned dabbling into this; if someone
could come up with a quantitative formula which is an improvement over
the intuitive, qualitative, by-guess-and-by-golly techniques that teams
now use, that would be a major contribution.
But, it'll be hard. There's not just the quality-of-competition issues,
but the context issues that we constantly mention on this list: players
roles on a team, coaches' strategic decisions, teammates' abilities and
roles. I doubt that any purely stat-based formula could have predicted
that Michael Jordan would become the NBA superstar that he did -- sure
he came from a good college program which played tough competition,
but there's dozens of players every year who come out of similar such
programs with similar stats. When NBA scouts saw future stardom for
Jordan, they were looking at things like his fundamentals (as DeanO
mentioned with James) and his physical gifts -- his explosiveness and
vertical. In other words, to use the baseball term, his "tools". What
marked Jordan apart from other college players was the very thing that
Beane in _Moneyball_ derides: looking beyond a players stats, at his
Of course what really marked Jordan apart was something even harder to
observe or quantify: his will to win, basketball intelligence,
dedication, etc. The sorts of personality characteristics that put a
Bird over a Barry, an Erving over a McGinnis, a Karl Malone over a Haywood.
Some of these can be observed (Lloyd Daniels though not a surefire bust,
had "questionable NBA future" written all over him), but most cannot.
At the SABR convention, a new subcommittee on "the science of baseball"
had some discussion about psychological testing and profiling of athletes
(NFL teams have done this for years, and baseball has done some of it).
But although that stuff has been going on for years in some sports, I
suspect that it's been only marginally useful, due to the large uncertainties
inherent in psychological profiling.