February 18th- Sports fans have always had a strange push-and-pull relationship with statistics: a lot of us feel an instinctive aversion to trusting numbers and formulas over what we observe with our own eyes; simultaneously, we love to have statistical evidence that backs our observations and theories about the games and players we love to watch and debate. To put it simply, we’re fickle and selective when it comes to stats.
Part of this fickleness stems from how vague and manipulatable stats can be. We’ve seen quarterback ratings and both 3rd and 4th down conversion numbers betray themselves as worthless out of context in the NFL; in baseball, many are finally coming around to understanding the limitations of stats like runs batted in, batting average, and wins; in NBA, instinct has told us that all 25-point-games aren’t created equal, that steals don’t validate good defense, and that no one who plays for (or against) a D’Antoni-coached team is as good as the numbers make it seem.
The NBA in particular is a nightmare of statistical manipulation, misinformation, and misunderstanding. There seem to be two extreme camps who dominate mainstream opinion: the Hollingers and Berris, who seem like stat determinists, employing formula-as-divine-truth; the stubborn traditionalists (too many to name) who point to “20 and 10”as a sacred cow, validate a single player’s worth by his team’s success, and bow to conventional wisdom as irreproachable (“Five out of seven NBA players I spoke to say Kobe is better than LeBron, so the stats can tell their story, but I’ll stick to mine”).
What serious NBA fans – we who knew Larry Hughes was never an All Star caliber player, that Jerry Stackhouse was less a burgeoning star than a man with a green light on mediocre teams, that Kevin Johnson was (and still is) the best point guard in Phoenix Suns history – want is a school of thought that bridges the chasm between the stubborn traditionalists and the stat determinists. Something to give us insight into why the stats that are thrown at us don’t always jibe with what we observe on the court; something that revamps and deepens the stat, so that it’s more comprehensive, more relevant to our understanding of the game. A form of statistical measurement that actually debunks and redefines other forms of statistical measurements. Something like… Sabermetrics.
So when a friend passed me this article from last week’s NY Times by the author of Moneyball,
I was excited.
At first look it seemed that Michael Lewis was going to bring to the NBA what he’d brought to MLB: advocacy for sensible usage of statistics; new models for team-building that bucked conventional wisdom; heretofore unknown, sophisticated methods to evaluate basketball talent and, in doing so, manipulate the free agent market. As I carefully read the entire piece, I felt initially confused, progressively indignant, and ultimately conned and offended. The future of basketball apparently lies in the deification of an unquestionable waste of a first round draft pick, and the General Manager shortsighted enough to trade away Rudy Gay for the services of the aforementioned older, proven-to-be-mediocre player. Instead of a profile of Billy Beane and the Oakland A’s, this was more like a celebration of Ned Colletti trading away young talent for Casey Blake.
“He was almost instantly dismissed, even by his own franchise, as a lesser talent.”
The centerpiece of Lewis’s profile is Shane Battier – a magnificent college basketball player who has averaged 10.1 points, 4.8 rebounds, and 1.8 assists during a far from illustrious 8-year career that the realistic among us would call, at our kindest, disappointing. Lewis (and Rockets GM Daryl Morey) acknowledges Battier’s athletic and statistical shortcomings, but only as a springboard to the (empty and warped) declaration that “every team he has ever played on has acquired some magical ability to win.” That such a marginal (by any measurable standard) player has been on winning teams is, in Lewis’ view, “a basketball mystery.”
The hard evidence to deepen the mystery, and support this bold link between Battier and winning basketball:
The Grizzlies went from 23-59 in Battier’s rookie year to 50-32 in his third year, when they made the N.B.A. playoffs, as they did in each of his final three seasons with the team. Before the 2006-7 season, Battier was traded to the Houston Rockets, who had just finished 34-48. In his first season with the Rockets, they finished 52-30, and then, last year, went 55-27 — including one stretch of 22 wins in a row.
Nevermind the fact that Battier started a grand total of one game for that 50-win Memphis team; that Pau Gasol was the obvious reason for their improvement; that James Posey, Jason Williams, and Mike Miller made such significant contributions; or that Lorenzen Wright and Bonzi Wells were as important if not more important to that team than Battier; Lewis manages to peg Battier as the root cause of the Grizzlies’ “magical ability to win.”
A person writing an article for The New York Times about advanced, sophisticated utilization of statistics might have engaged the cutting edge research method of google-searching the Houston Rockets 34-48 season: a season in which McGrady and Yao Ming played a total of 31 games together; a season that concluded with only seven players who started the season on the Rockets’ active roster. At the time the Rockets were – as anyone with a memory or a computer and a desire for such knowledge can ascertain – a season removed from winning 51 games. So a 52-30 record over the course of the ‘06-‘07 season was hardly unexpected, or evidence of a drastic turnaround. Nor can it be construed as proof positive that Battier carries with him the magical, mysterious smell, feel, and aura of winning basketball. He just happened to be there, and contribute a bit along the way. There’s not much magic or mystery to it – tagging along for a successful ride while contributing a little bit is the distinguishable hallmark of a role player.
(Somewhere, Devean George is waiting for his profile. To his credit, the Lakers actually won a playoff series with him on the team.)
That this piece has been met with such immediate and disproportionate interest and praise in the blogosphere, and that it landed in The New York Times Magazine is fairly disheartening to thoughtful sports fans. On the one hand, the buzz that has accompanied this story is encouraging because it indicates a general desire to embrace new methods of thinking about basketball and understanding how to evaluate players; on the other hand, the reaction betrays the general willingness to take as gospel or describe as inherently “fascinating”whatever new information is sold to us by someone with credentials. It’s a Bernie Madoff sort of scheme: the information promised is too grand, and the portfolios (Morey’s and Battier’s achievements), upon basic scrutiny, don’t quite support faith in the grandiose.
Am I the only one who finds it odd and disquieting that I sat down to read a story written by the man who helped get out the message that conventional statistics in baseball don’t tell the whole truth, only to find that he’s crudely manipulating and interpreting team win-loss records, taking them completely out of context, and using only scraps of questionable evidence to extol the virtues of Shane Battier and Daryl Morey? Am I the only one mystified by the resultant herd of people who are excited about this, and are recommending the piece to as many others as they can as a fascinating read?
Quantifiable Obama-ness Per Game
Beyond its drastic and unsupportable overstatement of Battier’s value, and Morey’s genius, Lewis’ piece has no idea what it’s about. Is it a personal profile? Is it an overhauling of statistics? Is it a scouting report on Kobe Bryant? Is it about the divide between “the streets” and half-black men like Shane Battier and Barack Obama? Because of that inherent confusion, it gives short shrift to each of the points it tries to unconvincingly (and sometimes unconsciously) advocate.
As far as statistical insight goes, Lewis’ piece simply doesn’t deliver. Unless you’ve been obstinate, willfully clueless, or just unaware for the past handful of years, you’re not getting anything new, interesting, or thoughtful there. The Sabermetric approach to statistics – expanding them, revamping them, and making more sense of them – has been outlined, explained, and rendered (comprehensively, intelligently) time and again. While it has almost exclusively pertained to baseball, the concept isn’t unfamiliar. NBA teams have been charting player tendencies and shooting % to the left, right, off the dribble for years – I knew this sort of information as a high school basketball player. That isn’t a revelation.
Lewis’ and Morey’s promising premise – that there’s a player in the NBA who appears to be mediocre, but is great in ways that current stats can’t appreciate – never bears out. It’s introduced, supported by a couple of questionable statistics, and then bolstered by the assertion that Battier is similar to Barack Obama and isn’t as flashy as street kids.
Even the analysis of the Rockets approach to defending Kobe is rife with dubious conclusions about Battier. Those who adhere to Sabermetrics would know that two games of defense against Kobe Bryant constitute a small sample size, and as such is inconclusive. And what to make of the fact that the defensively-challenged Golden State Warriors actually did a better job than the Rockets against Kobe last season?
Sadly, startlingly, the story has the most basic of seemingly innocuous racist leanings. Lewis relates an anecdote from Dan Wetzel that in high school Battier “was Barack Obama before Barack Obama.” Do we really need to analyze that declaration to peg it for what it is? Can anyone explain what that statement means? How it makes any sense? Why it’s germane to the story? Since it’s a physical and chronological impossibility, we’re left to infer that Shane Battier was the first articulate, principled, serious, black man Dan Wetzel ever knew; Barack Obama, years later, apparently was the second. While I’m sure (or, I should say, I hope it to be true) that is not what Wetzel would want to convey, I urge and challenge you to reread that quote and answer the questions I posed. It’s quite puzzling.
Why in a story about statistics, are we getting a David Eckstein sort of personal profile about grittiness and guts and smarts and lack of athleticism? As the piece progresses, I felt stuck in the middle of an unholy marriage of stats and anti-stats. Lewis and Morey seem to be saying: this player is valuable in ways commonly acknowledged stats can’t measure, but we can prove his value because the stats we’re using to measure it are nuanced, and innovative. Our stats? Intangibles, team win-loss record, mixed heritage, willingness to pore over reports on Kobe Bryant, lack of enjoyment of pre-game rituals, bizarre reluctance to shoot half-court shots at the buzzer, and disdain for the opinion of “the streets.”
By the end of this absolute mess an astute fan who came to the party looking for new measures of basketball value instead finds himself caught in an incohesive shootout between hackneyed generalizations about “the street,” articulate black men, Chris Webber, Kobe Bryant, and intangible goodness.
Why does Shane Battier’s personal profile matter at all? What can we glean from his white mother and black father, or from his tendency to withdraw from others, and how can we apply it to building a better basketball team? And what to make of this mythical Greek Chorus deemed “the street,” who collectively pronounced Shane Battier as unworthy of Chris Webber’s legacy? (Wait, so he actually is as good as Webber? Or is it that Battier is actually better than Webber because the Rockets rebound better with him on the floor? The points of logic throughout are so incongruous that it’s very difficult to know what’s being explicated.)
Ultimately, it seems a bit strange (if not incredibly brazen) to choose Morey and Battier as centerpieces in light of the fact that no sensible GM in the league would trade Rudy Gay for Battier, which is precisely what Morey did, and which runs antithetical to Billy Beane’s principles of trading. Instead of being taken to task for that deal, Morey is being deified for recognizing how the half-whiteness, suburban temperament, and sage mediocrity of Shane Battier makes for a better basketball team? Say what?
In the wake of digesting this piece I’m not even thinking about stats anymore, I’m thinking about how poorly conceived this whole story is. And, unlike those truly smart and comprehensive baseball pieces that elucidate the new methods of statistical evaluation of player worth, I don’t feel this piece left me with any greater understanding of some advanced, interesting method or analysis. Quite the contrary; I feel more muddled down in statistical manipulation than ever.
I’m left contemplating this philosophical maxim: truth does less good in the world than its false appearance does harm. As momentum builds behind Lewis’ truly weird hodgepodge of a story, my understanding of that deepens. To be fair, Lewis hints right off the bat that this might all be an elaborate ruse: “
Battier's game is a weird combination of obvious weaknesses and nearly invisible strengths." Take out the word “nearly,” and he might’ve been describing his own argument.
You can email us at feedback@tpfsports.com
