The IQ of Where to Be
Nerdy basketball – who knew? This is probably the coolest thing I’ve ever read about basketball – partially because of the subject, and partially because Michael Lewis is so good at writing about this kind of thing.
It reminds me a lot of a book I read a long time ago about Bill Bradley’s basketball career at Princeton called “A Sense of Where You Are”. Bradley had a somewhat similar approach to the game. The author, John McPhee, also knows how to put one word in front of another.
Here’s some other interesting work in sports economics on spillovers from teammates.
Kendall, Todd D. 2003. “Spillovers, Complementarities, and Sorting in Labor Markets with an Application to Professional Sports.” Southern Economic Journal 70:389-402.
Torgler, Benno. 2007. “‘La Grande Boucle’: Determinants of Success at the Tour de France.” Journal of Sports Economics 8:317-331.
Torgler, Benno and Sascha L. Schmidt. 2007. “What Shapes Player Performance in Soccer? Empirical Findings from a Panel Analysis.” Applied Economics 39:2355-2369.
Idson, Todd L. and Leo H. Kahane. 2000. “Team Effects on Compensation: An Application to Salary Determination in the National Hockey League” Economic Inquiry 38:345-357.
— Gabriel Rossman · Feb 17, 05:05 PM · #
Well, I’d say it’s a safe bet that you’ve spent a little time on this subject prior to my post!
What did you think of the NYT article (as an obvious expert)?
— Jim Manzi · Feb 17, 05:12 PM · #
Dating myself I’m sure, but IIRC, Sport Illustrated covered similar ground 20-odd years ago with a feature story on Kurt Rambis.
— Tony Comstock · Feb 17, 05:25 PM · #
Jim,
I’m not in a great position to judge as I’m neither a sports fan nor an active researcher on sports (I came across that literature while doing research on team spillovers in Hollywood). That being said, I think it’s entirely plausible, though with some caveats.
First, we need to respect the error term. There’s necessarily a lot more uncertainty about the claim that a particular player provides spillovers than the claim that spillovers are important in general.
Furthermore, unlike baseball, basketball and hockey are more than the sum of their parts which makes them subject to the Alchian-Demsetz theory of the firm. The only way to estimate spillovers is not a simple measure like RBI but a multilevel model with variance at the team and player levels. In order to disaggregate the two levels you need fairly high levels of trading AND you need a plausible assumption about efficacy over the player’s career length (i.e., at what age do player’s peak on average).
that assumes that spillovers are generic. if you make the horrifying but plausible assumption that spillovers rely on specific combinations of skills, then the problem becomes statistically intractable. since there are relatively low volumes of trading between teams the matrix of player combinations is sparse and the model is underdetermined. the only way to explore these assumptions would be some kind of agent-based modeling.
btw, you might be interested in some discussion of this at the sociology/OB blog orgtheory
Groupiness of Basketball
Groupiness in Sports Teams
gabriel
— Gabriel Rossman · Feb 17, 06:25 PM · #
Jim, That was an awesome article I would have missed otherwise. Thanks.
— Zak · Feb 17, 06:29 PM · #
Agree with Zak. Thanks, Jim.
Also, as an aside, this paper by Alchian and Demsetz bringing to the fore the “centralized contractual agent in team product process”, and, going waaaay back, Alchian’s economic survival principle, were among the first clues we had that Rawls’s original position was at least one distinct party short.
Their solution to the problem of “who will monitor the monitor?” is almost exactly the same as the living systems principle of vertical integration (vertical mutualism).
— JA · Feb 17, 07:50 PM · #
Jim,
In case you missed it, Steve Sailer also noticed this article and not surprisingly, had some interesting things to say:
http://isteve.blogspot.com/2009/02/rise-of-mulatto-elite-even-in-nba.html
— Jeff Singer · Feb 17, 11:36 PM · #
A surprising number of the top centers in NBA history have had three digit IQs. David Robinson scored 1300 on the SAT (old style scoring, equivalent to about 1400 today). Kareem Abdul-Jabar scored 1130 (old style). George Mikan was a successful lawyer after his playing days were over. Bill Walton attended Stanford Law School while out from the NBA with injuries. 7’-2” Dikembe Mutumbo got into Georgetown as a regular admittee to study diplomacy (and coach John Thompson just about had a heart attack when he saw him walking across campus with a stack of books in his arms — that’s why Georgetown had its unwieldy Twin Towers formation with Mutumbo and Alonzo Mourning — nobody had recruited Mutumbo). Wilt Chamberlain and Bill Russell always gave the impression in interviews of being sharp-witted men.
It’s possible that hard-working book-smart guys do better at center than at other positions. Playing with your back to the basket is rather unnatural, and typically requires extensive coaching.
— Steve Sailer · Feb 18, 05:09 AM · #
If you want to find intelligence on the floor, the obvious place to start is the point guard. More complex decisions are made at that position than anywhere else, usually without the benefit of physical size.
— JA · Feb 18, 05:38 AM · #
But, where’s the evidence that point guards have more book-smarts than centers?
I think we’re dealing with two different kinds of intelligence. Battier, who had a 3.96 GPA at suburban Detroit’s most expensive prep school, has the academic intelligence to devour detailed tables on opposing players’ statistical tendencies. The Rockets don’t bother giving the data to the other 11 players on the team. It might make them worse players, causing them to overthink.
— Steve Sailer · Feb 18, 06:48 AM · #
Sort of an exterior sense of proprioception.
— Julana · Feb 18, 12:26 PM · #
I guess the evidence is circumstantial. One thing we can agree on: the demands of position select for intelligence, to greater and lesser degrees (like your football post, where you showed how intelligence was unevenly distributed on the field).
Perhaps the distribution in basketball is not as pronounced. I don’t have any data one way or the other. But one position does stand out more than the others in terms of cognitive demand: the point guard position. For instance, that position selects for players who are exceptionally gifted at making rapid gestalt shifts among different “beliefs”. Not subconscious reactions, but affirmative judgments on many different levels of information, all happening in the blink of an eye.
Maybe not a difference in kind, since all positions demand similar baseline attributes, but definitely a difference in quantity and degree.
— JA · Feb 18, 03:25 PM · #
My hockey fan wife would also recommend Ken Morrow as a study in “the IQ of where to be.”
— Tony Comstock · Feb 18, 04:23 PM · #
Steve Sailor,
What Battier is described doing in that article is has been done forever by defensive minded players. Everyone has things they are less good at doing and places on the floor they are less successful. The very first rule of player specific defense is to force a player to his weaker hand. Rule no. 2 is make him take shots he is not good at. For instance, everyone knows you want LaBron James to take jump shots. And anyone who knows Kobe at all knows he will take as many bad shots as you give him. It requires absolutely zero amount of statistical research (or genius mulatto blood) to know that encouraging Kobe to take eighteen footers with a hand in his face is going to hurt his team. That has been Kobe’s MO for his whole career.
This has been an even less successful application of your obsessional theories than usual.
— cw · Feb 19, 02:23 AM · #
The point guard correlation would seem to make sense – a passing point guard processes a great deal of data very quickly, and I was a point guard, so I should be biased – but I doubt Battier-like book smarts are the key ingredient in great point guards. Perhaps something more like Gretsky-esque “physical genius” (to borrow from Gladwell)…something more fluid and intuitive. Jason Kidd, probably the greatest point guard since Magic Johnson (Stockton was a master in half court sets with a physically overwhelming pick-and-roll partner), is a local Oakland boy known for being a great and nice kid and also for possessing conspicuously modest intellectual endownments.
— Matt Feeney · Feb 19, 02:43 AM · #
Point guard is by far the most intellectually demanding position in bball.
— cw · Feb 19, 02:50 AM · #
To the extent that any of the traditional positions selects (at all) for intelligence, CW and JA are probably right, though again book smarts don’t seem to be what’s being selected for. (The greatest point guard in my own basketball world growing up, a true freak of on-court intelligence, was average, at best, intellectually. I simply have never seen the correlation in real life.) Re. Steve Sailer’s discovery of smart centers, I have my doubts. It may be that being freakishly tall tends to correlate with being detached and thoughtful, and also correlates with having had a decisive physical advantage at a time when coordinated seven-footers were rare in sports. (Or maybe the correlation is between being coordinated despite being seven feet tall and intelligence.) Kareem’s game was based on shooting the same unblockable shot 25 times a game. Wilt’s was based on being stronger and faster and more agile and also eight inches taller than his opponents at the position. Mutumbo was a smart guy but an extremely limited center. Robinson was distinguished less by court sense or anything else we might associate with intelligence than being 7 feet and having the physical wherewithal, the quickness and agility, of a six-footer. He was smart, but his freakishness was physical.
— Matt Feeney · Feb 19, 03:37 AM · #
I too was a point guard, and I embrace my bias.
Yeah, Matt, I’m not real sure what Steve means by book-smarts, because the intelligence we’re talking about is definitely not ‘verbal’. Spatial smarts, maybe? It is definitely associative, whatever we call it.
— JA · Feb 19, 03:58 AM · #
That seemed like an article by a layman for laymen. It was like a lot of popular science writing.
— cw · Feb 19, 04:00 AM · #
Maybe I’m too much a layman, but I thought Lewis used Battier’s story to get past the idea of “book smarts” and to explore what it means to live for aggregate results at the expense of individual samples.
— Matt Frost · Feb 19, 04:36 AM · #
Lewis used Battier’s story to get past the idea of “book smarts” and to explore what it means to live for aggregate results at the expense of individual samples.
Exactly, and something I’ve been pawing at for many years. If “the good” can be defined both horizontally and vertically, the collective has a vote, too.
The theoretical revolution in basketball happened when statisticians elected the self-interested perspective of the whole.
— JA · Feb 19, 04:51 AM · #
I think the article was best as a profile of Battier as a person. What I got from it is that he is smart, less athletic than most NBA players, and has used his smarts to figure out how he can play in the NBA. Bruce Bowen is probably not as smart but he figured out the same thing. That whole living for aggregate results is irrelivant. Playing that ways is the only way Battier can get paid in the NBA. He doesn’t have the physical ability to do anything else. His only option is to be what they call a glue guy, do all the littel unglamerous things that help a team win. LABron James has all kinds of options on the court and yet he also lives for aggragate results, but he doesn’t do it by blocking out the other teams better rebounder (which sounded like BS to me, you just block out who ever is closest). He does it by passing the ball, playing good defense, being a good teammate off the court, and scoring when it’s needed. There are lots of guys in the NBA who have the option of scoring more at the expense of winning but choose not to. That seems more interesting that Battier’s compelled altruism.
And what Battier is doing on the floor is good, but not revoloutionary or exceptional. He’s a fairly good defender who can hit the three and rebound and pass the ball moderately well without complaining that he’s not getting the ball enough. Those guys are fairly rare but they are out there. James Posey, Bruce Bowen to name a couple off the top of my head. Plus a bunch we don’t really know about. They are definitely vaulable on the right team, a team with people who can all do the other more obvious things. But Battiers play alone is not worth a article as far as I can see. I think his race and his infered protostant work ethic played a part on his selection as a subject. He’s appealing to a swath of white america that I suspect includes the author. They probably see him as a refreshing contradiction to the average NBA player. But I think that it would be more intersting to read about the average NBA player, because he is much less familier to us white folk than Battier, who is like David Watts from the KInks song, he meets the familier agreeed apon role model standards.
And finally, as I mentioned above, you don’t need reams of stats to figure out how to play Kobe, one of the most scruitnized player ever. That stuff about making him take a certain shot at a certain part of the floor is elementary basketball. TheRockets may have refined it, but to me it sounded like Morey was BSing the author about the magic of his propriatery stats. He’s probably BSing about Battier as well. Battier is this unrecognized all-star who I alone recognized and I have the stats to prove it. But the rockets are not an exceptional team in any way, so his stats haven’t had any effect on success that I can see, at least so far. Statistics are statistics in basketball and in social science. They can give you unexpected insights but they can also easily confuse the shit out of you.
Anyway, I can think of more interesting things to write about in the NBA.
— cw · Feb 19, 05:22 AM · #