http://www.bulletsforever.com/2014/7/25/5937399/john-wall-shooting-improvement-wizards-nbaNeil Greenberg at the Washington Post's newish statistics blog has an entry with the following headline: "Here's why Wizards would be better off if John Wall would stop shooting."
On the surface, that's absurd. Wall led the league in assists (tied for third in assists per game) last year and was third in the NBA in total passes per game, so it's silly to say, as Greenberg concludes, that Wall "has to think more about passing the ball than shooting it."
To be fair, the actual argument in the piece is a little more complex. Essentially: Wall is an inefficient scorer because he is not a good shooter, yet takes a lot of shots anyway. He also controls the ball a ton, which is well supported by SportVU data. Thus, "it appears Wall is hurting the offense" by dribbling to generate his own shot instead of passing it.
Still, this is a faulty conclusion, one that illustrates the complications of using any advanced sports data of this kind. These tools are often very good at displaying a problem -- Wall is inefficient and the Wizards offense also isn't great -- but very bad at providing a road map to fixing it. If basketball were as simple as Wall trading shots for passes, the fix would have already been made. Unfortunately, that's not the case. There are teammates, the shot clock, the time of the game, execution of the set that's called and (most importantly) the defense to consider.
Also, Wall's doing fine as is. Basketball is about people playing the part of what they should be, not what they actually are. Why do players inch closer to someone who has hit two shots in a row even though the odds say they are horrible shooters on average? Why does a player think they can hit a difficult shot more easily if he's just made two in a row than if he's missed? It's because of human nature. I stress this all the time because it's true: the threat of a player being capable of shooting is far more dangerous than their actual shooting percentage. That threat -- which incorporates more than just a single average number -- forces the defense to divert resources away from another potential threat.
In Wall's case, the threat of his shooting, whether improved or not, did make a difference. It provides more space to manipulate the defense and set others up for easier looks. Via NBAWowy, only Marcin Gortat (weirdly enough), Al Harrington, Drew Gooden and Garrett Temple had better true shooting percentages playing without Wall than with him. The Wizards as a team posted a 54.4 percent true shooting percentage with Wall in the game and a 51.7 percent mark with him on the bench. More fundamentally, their offense was 5.4 points better per 100 possessions with Wall in than out. If Wall's inefficient shooting really was detrimental to the point that it was hurting his team, those numbers would be different.
The reality: Wall, despite his inefficiency, is by far the biggest priority for a defense because of his speed, vision and, if we're being honest, the threat of an improved jump shot. He was by far the best playmaker on the Wizards, whether for himself or others. Replacing Wall with a lesser playmaker makes everyone worse, not better.