David Lee and Analytics: Separated At Birth
They say life comes full circle. Well, so do analytics.
Here at the MIT Sports Sloan Conference in Boston, everything from advanced stats, graphs, charts, tables, commissioners, reporters, general managers, coaches and anything in between congregate all in the name of sports analytics.
It’s a movement that has gained some legs over the past few years, but one that hasn’t really engaged the mainstream sports audience, yet. Known amongst the regulars as “Dorkapalooza,” Sloan’s main objective is “to provide a forum for industry professionals (executives and leading researchers) and students to discuss the increasing role of analytics in the global sports industry.”
There were numerous panels on Friday that were intriguing, including panels with Mark Cuban, Daryl Morey (GM of the Rockets), Stan Van Gundy, Paraag Marathe (COO of the 49ers), Bill Polian and Nate Silver. But there was one panel, one highlighting the deficiencies of David Lee of your Golden State Warriors, that trumped the rest.
The discussion, entitled “The Dwight Effect: A New Ensemble of Interior Defense Analytics for the NBA,” enlightened the room with new innovations using STATS (SportVu) technology. Based on a research paper written by Harvard scholars Kirk Goldsberry and Eric Weiss, the presentation, in detail, displayed David Lee’s ineptitude on the defensive end, especially within the interior.
The study shows that when David Lee is engaged with an offensive player and that player is within five feet of the basket — i.e. “close range” — he allows 53 percent field goal shooting — good for 51st in the NBA (between Jonas Valanciunas [52.8%] and Jordan Hill [53.9%]).
Goldsberry, who was moderating the panel, showed the visual above as well as some video footage (2/24 @ Milwaukee) to further support his results. The footage showed what Goldsberry referred to as “The Golden Gate” defense, which revealed Lee’s unwillingness to move when Bucks forward Derrick Williams drove into the lane. With Lee simply standing and watching, Williams was able to drive into the lane with little to no resistance.
Furthermore, the video described what Goldsberry called “The WTF Rotations,” which showed Lee in the middle of nowhere on multiple possessions either leaving the opposing defenders with wide open shots out on the perimeter or giving them easy strolls in the paint.
All the visuals drew laughs from the crowd, who may or may not have ever seen David Lee play basketball before. For myself, the footage was nothing new, as I see on a game by game basis what David Lee doesn’t do on that end of the floor. With that said, seeing the blood-like colors of David Lee’s graph compared to that of Larry Sanders was quite the eye opener.
It was especially interesting watching this presentation knowing certain members of the Warriors organization were in the room. Their collective hearts had to crumble when their prized possession, their lone All-Star representative and loyal cohort of owner Joe Lacob was made a mockery of right in front of their eyes.
The irony? Golden State is actually one of a select number of teams in the NBA that have invested in the SportVu technology — the others being the Rockets, Thunder, Spurs, Mavericks and Celtics. Whether the Warriors brass has seen this information (or that they even care) is another question, but the predictable summation of the panel: David Lee is bad at defense.
Where have we heard that before?
The panel was far from useless, in fact the analytical breakdown as to why David Lee is a bad defender was fascinating, but it just us another example as to why statistics and analytics still have a ways to go until they become a substantial force within the mainstream sports audience. Having additional information at hand is always useful, but whether the mainstream fan cares or can absorb that information to determine player tendencies is still the bigger question.
Mike Zarren, assistant general manager of the Boston Celtics, made this great point on the analytics process: whether you think you’re a “stats guy” or not, the age of analytical thinking is upon us whether you like it or not. In a back and forth with Stan Van Gundy during one panel, Zarren noted that although SVG doesn’t consider himself a “stats guy,” his thought processes and ways of studying the game are that of one.
This leads me to Friday night’s game against the Boston Celtics, also known as the “What Would Steph Curry Do As An Encore?” game. I was in attendance at TD Garden and it was in fact my first time covering a Warriors road game. Standing in the corner spot where Ray Allen made three pointers for five seasons was fun during my stroll pre-game.
The game itself was disheartening. Stephen Curry was cold throughout (6-for-22 shooting), Paul Pierce’s efficient night and the Celtics use of zone defense (along with Avery Bradley) dampened the Warriors hopes of winning the critical road game. The talk of the town, David Lee, also didn’t help with one of his worst games of the year (4-for-13 shooting, 10 points) despite grabbing 19 rebounds.
Lee’s game at TD Garden brought up another interesting aspect from “The Dwight Effect” panel. The idea that was brought up was of defenders choosing not to contest a shot in an attempt to pad their rebounding stats. It makes sense, given that the best rebounders in the league don’t necessarily correlate to the best defensive efficiency. Names weren’t given, but that surely sounds like a certain Warriors forward.
This theory, along with the previously explained interior deficiencies of Lee, were all visible during Friday night’s contest. Whether subtly or glaringly, the analytical breakdowns of Lee’s, well, breakdowns were clear. Add that on to Lee’s terrible offensive performance and you have yourself a bad game.
Lee, along with the rest of the team, looked slow, tired and rather beaten for the whole game, despite having a 54-53 lead with 8:01 left in the third quarter. The Warriors insurance policy all year, Jarrett Jack, also had one of his worst efforts of the season (1-for-9 shooting, two points, one assist), further proving that Jarrett Jack God Mode is nothing more than a rare anomaly instead of a dependent reality.
The Warriors losing aside, my experience at TD Garden was a memorable one. It’s always interesting, even being in the media, adapting and appreciating these new environments. Whether it’s Oakland, Houston, Boston or wherever, the ambience and aura’s of each are life experiences that you can’t take for granted.
While I may have not agreed with everything presented or said at the conference, I respect the information presented and have even greater respect for those involved with the conference. The future of analytics within sports is coming sooner rather than later, and for many that future is already here. It’s a process that many, including myself, are still getting used to, but it’s cases like “The Dwight Effect” that lead me to believe detailed statistics and analytics will have a broad, mainstream audience in the near future.