Paul Oberjuerge header image 2

Stat Wonks Discover NBA

February 24th, 2009 · No Comments · Basketball, NBA

Consider yourself warned, street-ballin’, gravity-defyin’ hoopsters. Stat geeks are descending on your sport, and the next thing you know half the fans in the country will be talking about how “you’re good but you ain’t no Shane Battier.”

Michael Lewis is the author of “Money Ball,” the book on Oakland Athletics general manager Billy Beane and his stat-driven analytics and how they enabled a team with limited resources to win the American League West regularly, the early part of this decade.

Lewis didn’t break any ground that modern baseball analysts, led by stat demigod Bill James, hadn’t been over before (a time or 50), but “Money Ball” gave it a name and exposed it to a wider audience.

And now Michael Lewis appears keen on showing that what appears to be the free-form game of basketball also can be broken down into little pieces and analyzed for some greater and more esoteric truth. His warning shot? This in-depth piece that recently appeared in the New York Times Magazine.

The gist of it: That the Brightest Guys in the NBA are now quantifying numbers that basketball, historically (and some would say “blessedly”), has ignored. And that Shane Battier of the Houston Rockets, who by traditional statistical measurements (points, rebounds and assists per game) is barely deserving of an NBA roster spot, is (in fact) one of the most valuable players in the league.

Who knew? The Rockets, apparently. And it set me to wondering which other NBA players might be equally as useful … without us knowing it.

Go ahead and read Lewis’s piece … I’ve got time. See you back here in 15 minutes …

Interesting essay, but annoying, as well, in that Lewis doesn’t clearly delineate what it is that the Rockets (and other teams, apparently) are measuring. (It’s a secret.)

He does hint at a statistic some media outlets have begun using, among them espn.com, and that is this: How a team fares while a player in on the court.

It isn’t unlike hockey’s plus-minus stat. If, during his time on court, a player’s team is outscored 72-60, he will have a plus-minus of minus-12. If his team outscores the opposition 80-75 while he is on the floor, he has a plus-5.

Then, apparently what we do is study several applications of this number.

–Is a player’s plus-minus out of whack with the old-school statistical barometers? To the good, like a Shane Battier? To the bad, like whoever the NBA’s answer to Juan Pierre is?

–Is a player’s plus-minus markedly above or below that of his team, as a whole? For example, if the Lakers are averaging plus-8.4 points as a team this season (and they are), and one player is significantly above or below that … something is going on that our usual metrics aren’t revealing to us. Especially this far into a season.

So, mulling this, I wondered, “Who is the Laker who gets the most minutes and seems to contribute the least?” And the answer to that pretty much has to be “Luke Walton,” does it not?

So, let’s check the plus-minus stat on Luke, and see if there is something going on here that the Lakers have identified and not told us about … or to see if Walton is as bad as traditional stats seem to suggest he is, meaning the Lakers are giving him significant minutes (and $30 million over six years) for no good reason.

OK, we may be on to something here.

Luke Walton is averaging 4.5 points, 2.5 rebounds and 2.5 assists in 40 games (he missed 17 because of injury), which isn’t much. At all. He makes 42.6 percent of his shots, which is below the team average, and makes 71.3 percent of his free throws, which doesn’t impact the team average at all.

He averages 16.9 minutes per game. Which is 35.2 percent of a 48-minute game.

Yet the Lakers are outscoring the opposition by an average of 3.9 points while Luke Walton is on the floor, and those 3.9 points represent 46.4 percent of the Lakers’ seasonlong rate of outscoring the opposition by 8.4 points. That is, nearly half of the Lakers’ “plus-8.4” comes during Walton’s barely one-third-of-a-game action.

If Walton were a merely neutral presence, neither helping not hurting in his 16.9 minutes per game, the Lakers ought to be outscoring the opposition by 3.0 while Walton is on the court. Instead, they are outscoring the opposition by almost a full point more during his time. A fairly significant number, it would seem.

Something is going on here. Could it be that Walton has been a starter for most of the season, and he’s being helped by playing a big chunk of his minutes with the Lakers’ starters? Well, maybe. Or is he doing some small, unmeasured things well, things new statistical models reveal to team executives?

This is worth more investigation. Not that I’m going to do it right this minute.

But here are some other stats to keep in mind: The Lakers are 47-10 this season. In games in which Luke Walton plays, the Lakers are 36-4, a winning percentage of .900. In games in which he sits out, they are 11-6, a winning percentage of .647.

So, anyway, perhaps Luke Walton has more value than we thought … and not only is he not as bad as we are inclined to think, he may be contributing to their success in ways that current statistics do not reflect. Whether that is defense, picking up loose balls, making the pass that sets up the pass … these numbers say the man is contributing something.

And, all you hoops fans, who thought stat geeks would leave alone your seemingly unscripted, free-form game … well, we may be just warming up, and the Michael Lewis piece was the numbers-crunchers throwing down a gauntlet.

Tags:

0 responses so far ↓

  • There are no comments yet...Kick things off by filling out the form below.

Leave a Comment