In the loss to Atlanta on Friday night, the Celtics’ fourth quarter offense generated some egregious shots on its way to 15 points, the most frustrating of those being a trio of Glen Davis mid-range jumpers (from 18, 21 and 21 feet), which always feel like sub-optimal selections, even when they splash down.
I wanted to have a look at the Celtics’ mid-range shooting percentages to verify whether or not we should be cringing when Davis jacks up a shot from this distance. As Henry Abbott alluded to in his internet-overloading post on Kobe Bryant’s clutch performance, our memories often prove unreliable indicators of actual performance.
The league average for mid-range shots (16-23 feet) is 39.4%.
Here are Boston’s mid-range shooters arranged by overall shooting percentage:
Delonte West: 53% (46% in 2009-10) on 1.5 attempts/game
Kevin Garnett: 46% on 5.0 attempts/game
Ray Allen: 45% on 3.4 attempts/game
Nenad Krstic: 44% (OKC) and 41% (BOS) on 3.2 attempts/game (OKC) and 1.6 (BOS)
Jeff Green: 40%(OKC) and 44% (BOS) on 2.2 attempts/game (OKC) and 1.8 (BOS)
Paul Pierce: 40% on 2.7 attempts/game
Rajon Rondo: 38% on 3.3 attempts/game
Glen Davis: 35% on 4.5 attempts/game
It’s damning evidence. Not only does Davis shoot the lowest percentage, he takes the second highest number of shots/game, after KG. In fact, you could reasonably assert that anytime the defense forces a Davis mid-range jumper, it’s a strategic victory for Boston’s opponent. Doc Rivers wants his players to take “their” shots, particularly if they’re open looks. The problem is that somehow the Celtics have determined that a Davis mid-range jumper is one of “their” shots, when almost anyone else on the floor, Rondo included, should be taking it before he does.
So, shouldn’t the Celtics redistribute shots from this range to improve offensive efficiency? The arguments against that are based in part on the idea that within the Boston offense, Davis will get far more open looks than, say, Allen or Pierce, because of the perceived offensive threats those players represent. At the Sloan Sports Analytics Conference this year, a presentation on optical tracking data suggested that tight defense (within three feet) drops expected shooting 12 percentage points. My evidence-free conclusion: Davis is much more likely to get shots free of these constraints than anyone else on the team but Rondo.
So, he should take them.
This is when we find ourselves down the analytical rabbit hole. If Davis is getting more open shots, shouldn’t he be hitting them at a far greater percentage than he does? Because we’ve already established that he’s one of the worst guys on the team from this distance, and below league average overall.
Which again suggests – fewer mid-range jumpers from Davis.
More on this topic tomorrow.