Nylon Calculus: On rotations, rest and ramifications for the Timberwolves
“I will have a Hall of Fame point guard on the floor at all times, I promise you that.
“I already know there will be times (when the game is close) and James will be saying, ‘This is bulls—t, I gotta be in there. And I’m gonna say, ‘I don’t know what to tell you. We talked about this.’”
The entirety of Jackie McMullan’s ESPN Magazine feature on the Chris Paul-James Harden trade is required reading, but the above quote from Mike D’Antoni was one of the most illuminating aspects in a retrospective reading, now more than a third of the way through the season.
As the dust settled on the landscape-altering trade, it may have seemed nothing more than a straightforward and innocuous statement from the Rockets’ head coach that he was going to stagger his two superstars’ minutes. But it is a little impressive just how true to his word Mike D’Antoni has been.
Chris Paul has rarely needed to play through the final whistle this season, a byproduct of a versatile and talented roster that is discarding opponents with relative ease so far. Paul is averaging 31.2 minutes per game, the lowest of his career, and James Harden is averaging 35.8, the lowest mark of his Houston career. The Rockets look determined not to repeat the mistakes of past playoff runs, as year after year, they made almost Sisyphean climbs only to look completely spent and collapse short of their ultimate goals.
The Rockets currently are everything that the Los Angeles Clippers, Paul’s former team, weren’t able to be. For years, the basketball world witnessed the Chris Paul-Blake Griffin-DeAndre Jordan triptych dominate opponents. For years, we saw Doc Rivers punish opposing teams with one of the most dominant starting lineups in the league. And across those same years, we saw the Clippers fail repeatedly to do much beyond that, beleaguered on a fundamental level by a lack of depth and an unwillingness to strategically optimize the talent that was available.
Earlier this year, Neil Paine of 538 did a quick little WOWY exploration of the Clippers’ stars. The results were predictable. The Clippers played really well with their Paul, Griffin, and Jordan playing together and fell apart when none of them were on the court, despite Rivers’ love for bench mob-style rotations and refusal to consistently stagger his stars’ minutes, which would have thereby at least attempted to mitigate the underlying lack of depth.
Lineups with Griffin and Jordan on the court together did reasonably well, a precursor to the Point Griffin style that we’ve seen the Clippers roll out this year, out of necessity though more than anything else. And while Blake Griffin and DeAndre Jordan ran the game well in their time together, it paled in comparison to the devastating Chris Paul-DeAndre Jordan two-man game. I’ve previously covered the reasons for that combination’s effectiveness, but that was a pairing that checked all three boxes of my lineup balance components: synchronousness, complementariness, and skill. It was almost coaching malpractice to push them to so many minutes as a trio when there was ample evidence that staggering them in pairs for longer stretches could have been advantageous — not only in the immediate goal of winning regular season games, but also in the long run with an eye on the postseason.
Andrew Johnson pointed out on Twitter that there was likely an underlying correlation between Rivers’ rotations and the Clippers’ inability to find an extra gear in the playoffs. He was responding to a comparison I had made regarding Doc Rivers’ substitution patterns. The team that I was comparing the Clippers to? The 2017-18 Minnesota Timberwolves, under the direction of an esteemed member of Doc Rivers’ own coaching tree, Tom Thibodeau.
Thibodeau, once a feared defensive mind who was near the top of the league coaching hierarchy, has become more infamous this season for his penchant for pushing players’ minutes too hard and running them into the ground. His bench players rarely get much run, and even worse, they’re not typically put in a position to succeed on the court, as Thibodeau has fallen in love with the bench mob rotations. The Timberwolves’ starting lineup of Jeff Teague, Andrew Wiggins, Jimmy Butler, Taj Gibson, and Karl-Anthony Towns is on pace to play over 1600 minutes this year and shatter the record for total minutes by a single lineup in NBA Stats’ database. The top nine 3-man combinations in minutes played this season are all various permutations of the Timberwolves’ starting five. Suffice it to say, this is not sustainable.
The problem is not simply that the Timberwolves’ starters play lots of minutes either. They also consistently play the longest continuous stretches of minutes in the league. Using play-by-play data, I tracked the average length of each stint for each player in the league this season. Four of the five Wolves’ starters were in the top 20, with Taj Gibson the lone exception coming in at 33rd overall. (There was some discussion on whether average stint length was more indicative than say, average longest stint per game, but the R2 between the two was a robust 0.84 so I left it as is.)
It’s not as simple as saying that players who play lots of minutes are bound to play the longest stretches of minutes as well, as the above top 20 is very much not congruent with the overall minutes leaderboard for the league this season. This is where smart rotations and the concept of strategic rest come into play. Staggering minutes can be an effective way to “artificially” reduce the burden on players throughout the duration of a game, beyond the perceived ham-fisted way of sitting players altogether (not that Thibodeau would ever consider that option either). As Tom Haberstroh documented late last year, DNP-Rest became a very real issue for the league, but smart coaches know how to buy time for their players inside of the actual games. And the Timberwolves’ starters could certainly use some.
"“We haven’t guarded anybody all year. We have to start guarding. It’s getting ridiculous.”"
Jimmy Butler ripped into the Timberwolves’ defense about a month ago, and why wouldn’t he? Defense is supposed to be the foundational identity of a team with Tom Thibodeau as its coach and Jimmy Butler as its focal point. Jimmy was slightly off though. It’s not that the Timberwolves couldn’t guard anyone; it’s that they couldn’t guard anyone in the fourth quarter. The Timberwolves have posted a 114.4 defensive rating in the fourth quarter this season, according to NBA Stats, the worst mark in the league. They have a semi-respectable 104.8 and 103.2 DRTG mark in the first and third quarters, respectively, which intuitively makes sense since those are the two periods when the Wolves’ starters are most rested. It all comes undone in the fourth quarter though as they simply run out of gas.
It would be one thing to suggest that the Timberwolves outside the starting five are simply unplayable which is why Thibodeau needs to push his starters as heavily as he does. But putting aside the fact that the talent on the team is still his responsibility, Nemanja Bjelica, Tyus Jones, and Gorgui Dieng all have posted a positive Box Plus-Minus so far this season! Jamal Crawford, for all his defensive shortcomings, has put up a positive Offensive BPM on the season. Granted, BPM is just one metric, but the point is this: the Timberwolves bench isn’t completely barren, and it certainly is not so deficient as to warrant 8-man rotations in December.
Next: Nylon Calculus -- Projecting the top 23 and under NBA players
Staggering en masse may perhaps not be the absolute salve for Minnesota. Wiggins is yet to live up to his once tantalizing potential, and Towns’ defense remains a confounding mess. There are real problems with this team, but the job of a coach is to optimize the hand that he’s dealt, and it’s hard to argue that Tom Thibodeau has done that up till this point. If he doesn’t want to follow in the doomed footsteps of his old mentor, Thibodeau would do well to take a few pointers from the coaches that he’s looking up at in the Western Conference playoff picture, particularly that juggernaut in Houston.
*Note: Data is recent as of 12/17. Problems with possession estimations used in ORTG and DRTG calculations by sites like NBA.com have been previously covered, however for the purposes of quarter by quarter comparison, they still provide an effective and quickly searchable data point.