Grading Minnesota Timberwolves star Karl-Anthony Towns’ 2019-20 campaign was a bit of a surprising exercise, as it turned out…
Most NBA advanced stats are made up of two components: a box component and a plus-minus component. In other words, “What did the player do?” and “How did the team perform when he was on the court?”, respectively. In this case, we’re examining Minnesota Timberwolves star Karl-Anthony Towns and his impact on the Wolves’ 2019-20 season.
For this reason, team-centric metrics like ESPN’s Real Plus-Minus (RPM) have a tendency to contradict others, and, at least when it comes to Towns, specifically FiveThirtyEight’s RAPTOR. Where RPM gives much more weight to the plus-minus than the box, RAPTOR values the box much more heavily.
In recent years, the box component for each has become much more advanced than that box component moniker might imply with the influx of player tracking data, to the point where, especially on the defensive end, traditional box score stats like steals and blocks make up only a fraction of the final equation. In RAPTOR’s case, blocks have been cut out entirely.
All this is particularly important when evaluating Karl-Anthony Towns’ 2019-20 season.
The Wolves were a bad defensive team with Towns on the floor in 2019-20, as has been well-documented. It was a major factor in his earning the second-lowest Defensive RPM among NBA centers at -1.58, and in turn, it became a subject of talking heads’ fascinations from around December through the potential end of the season in March.
It’s sad to see, really. Ask RPM’s leading scientist Kevin Pelton himself, and he’d likely give you a big asterisk on Karl-Anthony Towns’ low number. As part of his explanation of RPM, he says that “players can be helped or hurt by how well opponents shoot 3s and free throws with them on the court or on the bench – factors apparently largely outside of their control”.
That’s not going to determine the final number, but it can absolutely have an effect in extreme situations like Towns’ this year. Keep in mind Towns posted a DRPM of 0.84 a year ago and even improved from what was once a -1.92 this year.
Jump over to Ben Falk’s Cleaning the Glass, and you’ll see that Pelton’s disclaimer applies very specifically to Towns’ 2019-20 season. (Cleaning the Glass.com requires a subscription — or least a free trial, to see on/off stats.)
When Towns was off the court, Wolves opponents shot only 34.1 percent on 3-point attempts. When he was on, that number jumped up to 41.1 percent, a rather significant seven percent differential which placed Towns in the 1st percentile league-wide.
To attribute that completely to luck would be disingenuous, realizing Towns has had his defensive struggles, but there was certainly a bit of luck involved.
Towns’ biggest weakness has been in the pick-and-roll, and that didn’t seem to change in 2019-20.
But his his improvement as a rim protector which earlier in the year had fans and analysts largely optimistic that Towns was in fact turning a corner on defense, has gotten very little credit since.
To quantify this improvement, we turn to NBA.com’s player-tracking data which shows the players who forced the lowest field goal percentage on shots they defended within six feet of the rim. Early in his career, Towns’ tendency to chase blocks led to poor ratings in this statistic, but the last two years show a pretty significant improvement.
Among players who defended at least five shots per game within six feet (primary rim protectors), Towns has steadily climbed the charts and is now among the top third of qualified players (left side of the chart) in the field goal percentage he allows.
‘Season’ ‘Qualified’ ‘Towns DFG%’ ‘Rank’ ‘Towns DFG% Diff’ ‘Rank’
2015-16 38 56.1 26 -4.0% 24
2016-17 32 60.5 27 -0.3% 27
2017-18 35 61.5 31 -0.4% 31
2018-19 37 55.7 20 -6.9% 21
2019-20 50 51.2 16 -10.4% 13
Given the average field goal percentage of those who shot against him (right side of the chart), Towns was actually No. 13 among 50 qualifiers in decreasing that field goal percentage.
Sample size becomes a factor with Towns playing in only 35 games, but by all indications he’s gone from being a relatively useless rim protector in years two and three, all the way up to the top third in the league.
Returning to the discussion of RAPTOR versus RPM, RAPTOR’s biggest limitation on defense is that it has no way to measure individuals against the pick-and-roll. The pick-and-roll is still something that the team as a whole defends, but certain players like Towns make it harder for the team to excel.
RAPTOR’s strength is in assigning value to some of the other measures of defense, whether that be shots defended and points allowed, charges drawn, fouls committed, individual matchup points (how many points opposing centers scored), average speed, or a variety of other factors.
For centers, it values good shot blockers, and Towns is a good shot blocker.
So, which better represents Towns’ defensive impact? His Defensive RPM of -1.58?
… or his Defensive RAPTOR of 0.8?
The truth is that it’s a combination of both, with special attention given to the eye test as well. Towns did a lot of good things and made some notable improvements defensively this year. He really did. But he failed in other, probably more important aspects.
His improvements and non-improvements round out to about an average grade on defense, while his offense was simply spectacular.
2019-20 Season Grade: B
All condolences to the Towns family with the news of Towns’ mother, Jacqueline, passing in April due to complications from COVID-19.
Needless to be said, everything about that transcends basketball.