The NBA Finals and statistical bias


In an article for Grantland.com, Jonah Lehrer, author of “How We Decide” and contributor to Wired Magazine, pointed to this year’s NBA Finals between the Dallas Mavericks and the Miami Heat, a team loved by the numerically-inclined, as an example of the shortcomings of sabermetrics.

Lehrer, as one of the foremost thinkers in the scientific realm, writes that the Mavericks won the seven-game series, thanks, in large part, to Mavericks sparkplug J.J. Barea, who may have been hidden in the blind spot of sabermetrics.

Consider the case of J.J. Barea. During the regular season, the backup point guard had perfectly ordinary statistics, averaging 9.5 ppg and shooting 44 percent from the field. His plus/minus rating was slightly negative. There was no reason to expect big things from such a little player in the playoffs.

And yet, by Game 4 of the NBA Finals, Barea was in the starting lineup. (This promotion came despite the fact that he began the Finals with a 5-for-23 shooting slump and a minus-14 rating.) What Dallas coach Rick Carlisle wisely realized is that Barea possessed something that couldn't be captured in a scorecard …

… Because nothing messes with your head like seeing a guy that short score in the lane. Although Barea's statistics still look pretty ordinary — his scoring average fell in the Finals despite the fact that he started — the Mavs have declared that re-signing him is a priority. Because it doesn't matter what the numbers say. Barea won games.

While I whole-heartedly agree that Barea’s promotion had an impact on the series, I am hesitant to believe that his insertion into the starting lineup was a move that was not fueled by analytics.

Why? Because of a guy named Roland Beech. As the founder of the indispensable basketball analytics warehouse 82games.com, Beech was hired by Mavericks owner Mark Cuban last year to inform the coaching staff with statistical insight that could optimize, among other things, the lineups that the Mavericks employed.

Beech is some form of a pioneer. As far as I can tell, he’s the first sabermetrician (I use that as an pan-sport term because the basketball version “apbrmetrics” doesn’t exactly roll off the tongue) to sit on the bench of a major sport as a member of the coaching staff. And now, it’s not just any coaching staff – one that led the Mavericks to a title.

Take it from Cuban: Beech has influence. In an article by ESPN’s John Hollinger following the Mavericks’ Game 6 win over the Heat, Cuban spoke about Beech’s instrumental role.

“Roland was a key part to all his,” Cuban said. “I give a lot of credit to Coach Carlisle for putting Roland on the bench and interfacing with him, and making sure we understood exactly what was going on. Knowing what lineups work, what the issues were in terms of play calls and training.”

It’s also worth highlighting an SI.com interview that was published Tuesday with Dwane Casey, the Mavericks assistant coach before recently taking the Toronto Raptors head coaching gig. Casey was introduced to advanced statistics in Seattle several years ago when he worked with Dean Oliver, who is now ESPN’s director of analytics. At SI’s Point Forward blog, Casey explains the history behind his relationship with advanced statistics:

I didn’t really understand the value of those numbers at that point, even though Wally Walker [a longtime Seattle executive] was really pushing us to embrace it. But being from the old school of coaching, I didn’t really know what to do with all of it.

And in Minnesota, we just didn’t have anything like that. But in Dallas, Rick Carlisle and Mark Cuban have embraced analytical numbers, and they have been so valuable to us in terms of what lineups we put on the floor. And I know Toronto already has guys in place who do this. This is the new wave in the NBA.

Although we can’t know for sure that Beech was the brains behind Barea’s insertion into the lineup, we know he’s in the middle of every on-court personnel decision. After all, this is why he’s on the staff. Sabermetricians shout the hazards of small sample size so it makes sense that Beech and the Mavericks wouldn’t lend too much credence to Barea’s minus-14 plus-minus over a three-game sample or his lukewarm field goal percentage. The statistics that the Mavericks use are far more sophisticated and granular than the stuff we see on the television or on the internet.

In reality, the Mavericks’ victory should not be viewed as a win for the statistical skeptics since Cuban has put together one of the most analytically-savvy staff in sports. Advanced statistics and lineup data go hand-in-hand with the Mavericks’ decision-making.

Lehrer drew an analogy of horsepower and cars to illustrate the dangers of statistical devotion. Everyone talks about horsepower even though it’s not what really drives customer satisfaction. Lehrer elaborates on this point:

The explanation is simple: The variables don't matter nearly as much as we think. Just look at horsepower: When a team of economists analyzed the features that are closely related to lifetime car satisfaction, the power of the engine was near the bottom of the list.

Ah, but there is horsepower in baseball. It is batting average. And there is horsepower in basketball, too. It is scoring average. For decades, the sports world has bowed before batting average and scoring average (points per game) as the statistical standbys in their respective sports. You could toss RBI and field goal percentage in there too.

To use a line from Lehrer, they don’t matter nearly as much as we think. Sabermetricians have discovered better ways to measure a player’s worth – stats such as on-base percentage (OBP), weighted on-base average (wOBA), player efficiency rating (PER), and wins above replacement (WARP). Knowing that there are other important ways to contribute to wins, these advanced statistics try to expand the spectrum of analysis, not shrink it.

I have not met a sabermetrician who believes that there is nothing to be gained outside of the realm of analytics. Intangibles exist and they have power, even if sabermetricians have struggled to pinpoint and measure that power. No sabermetrician would honestly believe that they are all-knowing. That is precisely why they have strapped themselves in for this quest for objective information.

The Finals should be celebrated, among other things, because we were watching two organizations that successfully blended analytics and scouting into their winning programs. Heat head coach Erik Spoelstra actually built an advanced statistics database from scratch years ago for his boss Pat Riley when Spoelstra was an assistant. Carlisle wanted a stat geek on his bench in order to be better equipped to do his job.

If this is the new wave in the NBA, the Heat and the Mavericks are rolling through the league one win at a time.

UPDATE: Looks like Lehrer responded to my response to his Grantland article. Check it out at his blog where you'll find an interesting back-and-forth with some of his readers. As Lehrer said to me on Twitter, "This is fun."