The Dwight Effect: Where did Superman’s defense go?

By: Jamal Salmon

“I want to thank my teammates for allowing their men to get to the basket again, forcing me to block shots,” Dwight Howard joked as he collected his third consecutive Defensive Player of the Year award following the 2011 season, trying to give his teammates at least some credit for his defensive dominance. That season, Howard’s Orlando Magic miraculously posted a 98.9 defensive efficiency rating, good for 3^rd in the NBA, despite a roster made almost exclusively of offensive-only players. Infamously inept defenders such as Ryan Anderson, Gilbert Arenas, Hedo Turkoglu and J.J. Redick logged big minutes every night and somehow the Magic were still top 3 in the NBA in defense that year, as they had been each of the previous 3 seasons. Howard’s superhuman efforts on the defensive end were nothing short of spectacular, earning him the well-deserved, even if self-proclaimed, nickname of ‘Superman’. Frankly, anybody capable of lining up next to Rafer Alston, Vince Carter, and Rashard Lewis and still getting defensive stops can call himself or herself whatever they want. For Coach Stan Van Gundy life was easy. On most nights, he could focus solely on putting the best offensive players around the big man, knowing that Howard would take care of the defense – all by himself.

At the time, I remember referring to this luxury as a product of the Dwight Effect. Simply put, the Dwight Effect (D.E) are the points saved per 48 minutes by the impact of Howard’s presence on his team’s ability to play half-court defense. More precisely, the D.E measures Howard’s ability to decrease his opponents’ expected points per shot attempt (exPPS), both directly, by blocking and changing shot attempts at the rim, and indirectly, by deterring close shot attempts in the first place.

 

Calculating Expected Points per Shot (exPPS)

In order to explain Howard’s effect on opponents’ expected points per shot, it’s important to walk through both what exPPS measures and how it is calculated. Expected points per shot attempt is an estimation of what an offense can expect to score with each additional opportunity to shoot at the basket. It takes into account both the shot location and success probabilities and the associated point values. The shot locations are within the restricted area (R), otherwise in the paint (P), mid-range (M) and three-point range (T).

Expected Points per Shot (exPPS) = 2[R_f(R_%) + P_f(P_%) + M_f(M_%)] + 3[T_f (T_%)]

(where X_f = percentage of shots from location X and X_%= shooting percentage from location X)

With an exPPS of .9815, this sample team can expect to score .9815 points for every shot it takes during a game. Defensively, in order to lower exPPS, one must force their opponents either to take less shots at the more efficient locations on the floor or lower their opponents’ shooting percentages at these efficient locations.

 

Superman in Orlando

Coming into this season, no one in the NBA over the past four seasons has been better than Dwight Howard at doing those two crucial things on defense. The chart below shows the difference between his opponents’ exPPS when Howard was on the floor versus when he was not. Additionally, the chart shows the Dwight Effect, or how many points Howard’s defense alone saved his team that season per 48 minutes.

From 2008-2012, Dwight Howard was so dominant on defense that he saved his team on average 5 points per 48 minutes. To put that in perspective, over a full season that would be enough to vault a .500 team playing at a league average pace from 41 wins to over 57, according to Pythagorean win expectancy calculations.

 

Cape-less in L.A

Considering Dwight Howard’s history of carrying below-average defenders to above-average team defensive outputs, concerns about the Lakers inability to keep defenders in front of them were generally ignored. In fact, many believed that finally Steve Nash’s inability to play defense would no longer be an issue with Howard waiting in the wings to block every shot. Unfortunately, Howard has not been the dominant interior defender he was in Orlando. However, news hasn’t yet caught on around the league, as Howard’s presence still decreases shot attempts around the rim by more than 6 percent for the third consecutive year. While Howard’s reputation still has many would-be attackers settling for mid-range jumpshots, those who have attacked the rim have had shocking success with Howard in the game.

So far this season, opponents have shot 2.6% better in the restricted area with Howard manning the middle than with him on the bench. This is the first time since the 2008-2009 season that Howard has seen opponents have more success at the rim with him on the floor than with him on the bench.

Howard’s struggles to protect the rim (at least relative to his superhuman standards set in Orlando) and unusually hot perimeter shooting by opponents while Howard has been in the game now paint a grim picture of the Dwight Effect.

By no means, does this data mean that Dwight Howard is no longer an impact defender. An important issue to keep in mind with these numbers is that the season is still relatively young and as the sample size grows Howard will be less penalized for the unsustainably hot outside shooting of his opponents. I do believe that this data validates the claims of many around the league that Dwight has not fully healed from his back surgery and seems to lack some of his former athleticism. There appears to be a consensus that the Dwight Howard of old will return some time this season and, for a Los Angeles Lakers team struggling to get stops on a nightly basis, Superman’s return cannot come soon enough.