Rudy Gobert’s name is synonymous with elite defense.
The towering French center for the Utah Jazz is averaging at least two blocks per game for the sixth straight season. The Jazz are currently ninth in the league in adjusted defensive rating, their worst finish since Gobert became a full-time starter. Ever since entering the NBA, Gobert has played defense at the highest level and has transformed the Utah Jazz on that side of the ball.
His performance certainly has not gone unnoticed either. Gobert has been named to three consecutive All-Defensive First Teams and will likely be selected to his fourth this season. He is already in legendary territory as one of only ten players to win multiple Defensive Player of the Year awards¹ and could very well win a third in 2020.
Needless to say, Gobert’s status as an all-time great defender is not in doubt. However, we have access to more than just accolades and box score stats. There are more advanced defensive metrics out there than ever before. Do these metrics back up Gobert’s legendary defensive status? Probably, but let’s confirm it.
The first defensive statistic that one may think to look at is a player’s field goal percentage allowed. Just like offensive FG%, though, FG% allowed is flawed because it treats every shot equally without accounting for shot distance and shot value (two-pointers vs three-pointers). That’s where Expected Effective Field Goal Percentage Allowed (DXeFG%) comes in. A player’s DXeFG% represents the eFG% a league-average player would be expected to give up if they defended the same exact shots. Guards tend to have lower DXeFG% values because they guard other guards who shoot more jumpers, while big men have higher DXeFG% values. A player’s DXeFG% can then be compared to the actual eFG% they allow (DeFG%) to see if they perform above or below expectation. For more on DXeFG%, you can read our introductory article on it here and access the data via the site’s menu. Anyway, here’s a bubble plot of every qualifying player’s DXeFG% and DeFG% since 2017.
As you can clearly see, his point is far below the expectation curve, meaning he allows an eFG% far below expectation. Based on the shots he defended, Gobert was expected to allow an eFG% of 52.8%. Instead, defenders boasted a 45.8% eFG% against him. The sample size isn’t paltry either. Since 2017, Gobert has been credited with defending 4542 field goal attempts, the most in the league during this span by a wide margin.²
When DeFG% is used in conjunction with DXeFG%, it’s a far better statistic than regular FG% allowed (or DFG%). However, it’s still quite imperfect. I previously mentioned that DFG%’s issue is that it assumes all shots are equal. While DXeFG% offers a solution to that problem, it still makes one faulty assumption — all offensive players are equal. Holding Stephen Curry to a 3P% of 35% is pretty damn good, while allowing that same 3P% to Giannis Antetokounmpo is a bit disappointing. In the eyes of DXeFG%, both players are equal. That’s not right and it penalizes players like Jrue Holiday who are consistently matched up against the opposing team’s best offensive player.
Another one of our defensive metrics is based on individual matchup data. It takes a look at every matchup a defender has had and determines an expected ‘points allowed’ value based on the equality of that offensive player. Because we know how many points a defender actually allowed in any given matchup, we can determine how many total points each defender saved (TPS) and the number of points a defender saved per 100 possessions (DPS). Check out the introductory article here for more, and once again, full data can be accessed via the site’s menu under the ‘NBA’ tab. Let’s see how Gobert’s DPS ranks in the league since this individual matchup data began to be tracked in 2018.
According to the league’s tracking data, Gobert has defended an opposing player on 8024 total possessions since the 2017–18 season. It’s not the most in the league (as the graph clearly shows) because guards defend ball-handlers more often than big men. In fact, Gobert is 36th in most plays matched up against an offensive player since 2018. Nonetheless, he has saved 762.5 total points, the most in the league during this span. Forget volume, though. Nobody comes close to Gobert’s efficiency either. The ‘Gobert Line’ represents his DPS: total points saved per 100 possessions. Since 2018, Gobert has saved 9.50 points per 100 possessions — also the most in the league in this span.
So, Gobert dominates the competition in both of The Spax’s original defensive metrics. There’s one other way we can look to evaluate Gobert’s defense, though — value to a team. Both DXeFG% and DPS have to do with individual defense. One is based on an opponent’s eFG% when defended by a player, while the other is based on point production based on individual matchup data. In both instances, the key is individual matchups. How can we make sure this production is actually valuable to the team as a whole? Surely it is, but what if there’s players who are more valuable without putting up those gaudy DXeFG%/DeFG% and DPS numbers?
Value can be quantified through the use of Defensive Player Impact Plus-Minus (D-PIPM). PIPM is similar to other plus-minus metrics like RPM and RAPM in the sense that it builds upon traditional plus-minus (+/-) by introducing a prior and taking factors such as lineup strength into account. PIPM was created by Jacob Goldstein and you can read about it in detail here.
Instead of plotting cumulative D-PIPM, the plot below includes a point for every single-season performance since the 2009–10 season. Gobert’s seasons are clearly highlighted.
Can’t find Gobert’s points? Just ignore the bottom 98.9% of points. and it might become easier. The lowest D-PIPM Gobert has ever put up was 3.24 in his rookie season when he was averaging just below ten minutes per game. That’s the 55th highest D-PIPM recorded since 2010 out of 4853 total single-season performances. In other word, Rudy Gobert’s least valuable single-season defensive performance was still more valuable than than nearly 99% of all single-season defensive performances since 2010. That’s not too shabby. The last time he wasn’t in the top-2 was in 2016.
Yeah, Gobert doesn’t just put up great individual defensive statistics. He’s the most valuable defender in the game.
But why? Gobert has “only” led the league in blocks once yet he’s clearly the league’s premier defender. What exactly makes him so special and valuable? What puts him ahead other rim protectors?
The Utah Jazz employ a defensive strategy that is entirely built around Gobert’s dominant presence inside the paint. They look to run offensive players off of the three-point line and force them inside where they’ll be met by Gobert. This tactic can be seen in the stats:
Since 2017, the Jazz have given up the least three-point attempts in the league by far. Of the three-point attempts they do allow, 47.6% of them are classified as ‘wide open’ (nearest defender 6+ feet away) by the NBA’s video tracking system. That’s the fourth lowest rate in the league. A large part of Utah’s focus is clearly on limiting quality looks from beyond the arc.
You would think that this means Utah’s defense is spaced out which lets opposing players drive into the paint with ease. That’s where Gobert comes in. We’ve already seen all of his monster individual defensive numbers. The Jazz just want to put him in that situation where his skill set can be put to use.
In this play, Luka Doncic is forced to drive inside on Gobert. The Jazz provide absolutely no help defense which would allow Luka to kick it out for an open shot. Gobert doesn’t need it. Just leave him on an island and let him go to work.
- Dikembe Mutombo (4), Ben Wallace (4), Dwight Howard (3), Sidney Moncrief (2), Mark Eaton (2), Dennis Rodman (2), Hakeem Olajuwon (2), Alonzo Mourning (2), Kawhi Leonard (2), Rudy Gobert (2)
- The difference between Gobert and second-place (Karl-Anthony Towns with 4173 DFGA) is approximately equal to the difference between second-place and seventh-place.
Originally published at https://www.thespax.com on February 5, 2020.