Nylon Calculus: Missed 3-pointers are ruining Kawhi Leonard’s defensive numbers

Mar 9, 2017; Oklahoma City, OK, USA; Oklahoma City Thunder guard Victor Oladipo (5) shoots the ball over San Antonio Spurs forward Kawhi Leonard (2) during the first quarter at Chesapeake Energy Arena. Mandatory Credit: Mark D. Smith-USA TODAY Sports
Mar 9, 2017; Oklahoma City, OK, USA; Oklahoma City Thunder guard Victor Oladipo (5) shoots the ball over San Antonio Spurs forward Kawhi Leonard (2) during the first quarter at Chesapeake Energy Arena. Mandatory Credit: Mark D. Smith-USA TODAY Sports /
facebooktwitterreddit

Kawhi Leonard is the reigning two-time Defensive Player of the Year. He is known as one of the best perimeter defenders in the game. His nickname is The Claw.

So why are the Spurs so much better on defense with Kawhi off the court this season?

With Kawhi Leonard on the court, the Spurs’ defensive rating of 103.7 points per 100 possessions is not at all bad. In fact, if that was their rating for the entire season, it would be fifth-best in the league. So not too shabby at all.

But with Kawhi on the bench, the Spurs have had an unbelievable defensive rating of 95.5 points per 100 possessions. Removing the Defensive Player of the Year improves the defense by 8.2 points per 100 possessions. That seems quite strange.

Read More: Paul George is settling for bad shots

There are a multitude of factors in play here and one of the big ones is 3-point shooting. Justin Willard has pointed our here that opponents have hit the highest percentage (37.2 percent) of 3-point attempts with Kawhi on the floor of all players on the Spurs. And to make matters worse, opponents have also hit the lowest percentage (28.1 percent) of three-point attempts with Kawhi on the bench, also noted by Kevin Pelton on ESPN Insider.

This matters because, as we have previously written about here on Nylon Calculus, opponent 3-point percentage is mostly random across a large enough sample size. Teams have very little control over the opponent 3-point percentage. They can affect the number of attempts, but not the percentage. This means that there is very little Kawhi Leonard can do to affect the 3-point percentage of the opponents. The number of 3-point shots taken with Kawhi on/off the court is almost the same (24.4 vs. 24.9 per 100 possessions), so that it not part of the issue.

The league average 3-point percentage is 35.9 percent. If we pretend that Spurs’ opponents had shot league average on all 3-point shots, Kawhi Leonard’s on/off defensive rating improves by 6.7 points per 100 possessions. It is especially his off numbers that improve by quite a lot, because opponents have shot so poorly against his teammates:

That 6.7 points per 100 possessions seems like quite a lot, because it is. For comparison, it is the distance from the best defense in the league (San Antonio) and the 21st-best defense (Minnesota).

Comparing Kawhi’s correction with all other players in the league, we see that he is quite the outlier (improvements are negative, since you want a low defensive rating):

Kawhi has the biggest improvement of all 234 players that have played more than 1000 minutes. Only two other players come close. Deron Williams (in his minutes on the Mavericks) with an improvement of 6.3 and Andre Drummond with an improvement of 6.0 points per 100 possessions. Drummond is another guy who’s defensive on/off numbers have been terrible this season, but it is in part because of how well opponents have been hitting 3-pointers with/without him (31.7 percent vs. 39.5 percent) on the court.

The top three on the other end is: Joe Harris (+5.9), Jonathon Simmons (+5.8) and Thaddeus Young (+5.7). Simmons makes sense, since he plays a lot of minutes when Kawhi is on the bench, so he gets the benefit of those bad-shooting opponents.

Fun fact: If I lower the minutes limit to 500 minutes, there is actually one player with a bigger improvement than Kawhi Leonard. Nerlens Noel, in his time on the Sixers, gets an improvement of 7.8 point per 100 possessions. Noel’s on/off defensive numbers looked quite bad during his time this year on the Sixers, but that was in part because opponents with him on the court hit their 3-point shots at 44.5 percent (!). Kawhi’s outlier status this year actually holds across multiple seasons.

This is definitely not the end of the line for the kind of thinking. The estimates could be improved. Some of the shots that would have been made or missed would be offensive rebounds, so the offense would score anyway. One could adjust for that and it would likely make the distribution above slightly more narrow. One could also look into the lineup data to figure out what opponents were actually on the floor and use their 3-point percentages instead of league average.

It can seem strange that 604 3-point shots against the Spurs with Kawhi Leonard on the bench is not enough for the percentages to regress to around league average. But even though 604 shots is more than enough to make most of us worn out from shooting them, it is not a big number in a statistical sense. Give a league average (35.9 percent) shooter 604 attempts and around three out of 10 times he will shoot either above 38 percent or below 34 percent, which would probably lead us to call him a good shooter or a bad shooter.

Next: Nylon Calculus -- The curious case of Boogie and the Brow

Kawhi’s defense has not been bad this year. His team has been really good on that end with him on the floor. They have been even better with him off the floor, but that is in part because of the horrendous shooting of opponents. So Kawhi can look to the other bench and blame those guys for making his defensive numbers look bad.