Statistical Analysis

Putting the Timberwolves Minutes Controversy in Context

 

The Timberwolves starters are playing a ridiculous amount of minutes this season. What does that look like when put in context?

(Editor’s Note: The statistics below are current through December 15. The W-L numbers and conference standings are current as of December 21.)

Tom Thibodeau is known for riding his starters hard, perenially seeming to be the coach whom people free associate with unnecessarily arduous minutes for his main rotation players. Although Thibs’ proclivity to play his starters long minutes was far from a secret when he joined the Wolves before last season, an increasing number of people—national writers, local writers (Editor’s Note: AWAW recently did a massive roundtable discussion on this issue and you really need to read it), hordes of frustrated Wolves fans on Twitter and Canis Hoopus comment threads, and others—are saying that enough is enough: Thibs needs to cut back the workload of his starting five. They might get tired in the 4th quarter. They might face a heightened injury risk. They might be worse on the tail end of back-to-backs.

Or, according to others, the real reason Thibs’ starters have played so many minutes this season is due largely to a lack of depth. POBO Thibs improved the roster from last season in an attempt to optimize it for wins, and not for wins in a few years—Thibodeau wanted to win this season.

You can make an argument that he’s right. The Wolves haven’t made the playoffs since 2003-04. For those joining us late, it’s now 2017. Unless there’s a nuclear holocaust involving North Korea in the next 10 days—never say never—the calendar will roll over to 2018 soon.

Kim Jong-Un yukking it up with former NBA legend Dennis Rodman.

And a lot of us want to win, and win now, and if the minutes pile up, well, at least we tried.

One can argue that so far, this approach is working: the Wolves are currently 19-13 (a .594 winning percentage), which is good for 4th in the always-competitive Western Conference. Although no one knows what the future holds, what is obvious is that the Wolves are not as bad as some of the most vocal Thibs critics, who are calling for his firing, suggest. At their current pace this season, the Wolves are on track not only to make the playoffs for the first time in over a decade, but also to win almost 50 regular-season games before enjoying home-court advantage in the first round of the playoffs.

You probably don’t need a reminder about how truly abysmal the Wolves have been for years and years and years. But here’s one anyway. To make a long story short, Timberwolves basketball has been synonymous with losing. Did you get excited for a second like I did remembering that the 2013-14 squad flirted with a .500 record?

I did too. Bad, bad, bad.

How We Got Here

The 2014-15 team—which was 16-66 in Flip Saunders’ last full season as Emperor and Grand Poobah of the Wolves—tanked its way, along with some good ping-pong ball luck, to the number one pick in the 2015 draft. Thusly, a KAT became a Wolf. Kevin Love got tired of losing and wanted out. The Wolves dealt him to Cleveland. Thusly, Andrew Wiggins became a Wolf. (Editor’s Note: The Wolves also got former number one overall pick Anthony Bennett from Cleveland—we still love you AB fuck the h8erz.) As we know, the Wolves burst on to the national scene as a possible contender with the 2017 draft-night trade with Chicago, which brought us Jimmy Butler. Thibs filled out the starting lineup by signing Jeff Teague and Taj Gibson to replace Ricky Rubio and Gorgui Dieng.

To supplement the starting unit, POBO Thibs also re-signed Shabazz Muhammad to a minimum deal in free agency, but only after Bazz realized he had little value or bargaining leverage on the open market. It was considered to be a good deal for the Wolves at the time.On his affordable contract, Bazz (Editor’s Note: Does anyone know when Bazz is planning to go through with the legal name change so we can just refer to him as Bazz without first invoking “Shabazz Muhammad”? He feels like “just Bazz” already.) was supposed to provide wing depth and receive sufficient playing time, such that he would be able to help the team with scoring and energy off the bench whilst increasing his value for another run at free agency after the season. But Bazz quickly played hisself out of Thibodeau’s rotation, which was shortened to eight players after Thibs shelved Bazz on November 29. He remains out of the rotation, recently replaced by Marcus Georges-Hunt.

Equally troubling from a depth perspective is the absence of Nemanja Bjelica, another key rotation player. After two less-than-stellar seasons with the Wolves, Belly, who was awarded the 2015 Euroleague MVP, appeared to (finally) be breaking out as the kind of invaluable bench player other teams can only ignore at their peril. When Bejelica went down with an injury in November, he was leading the NBA in three-point field goal percentage. Perhaps equally important, he was a threat on dribble penetration to score or kick the ball out to open teammates for easy catch-and-shoot opportunities.

The Wolves miss “Good Belly” in a bad way. Bjelica took a lot of pressure off the rest of the team with his style of play and caliber of performance off the bench. Belly, who has missed the last 14 games with an “increasingly mysterious ‘mid-foot sprain,’”, is working out again but is expected miss more time.

So yeah. The Wolves bench is essentially shot. We haven’t even gotten to Jamal Crawford yet, but an entire other post could be written about his travails despite his vastly improved play this week. (Editor’s Note: Britt Robson has already written it over at The Athletic.) Tyus Jones and Gorgui Dieng have been the only steady bench performers. Neither can play the wing. For the 2017-18 Timberwolves, depth at the wing is kind of like the Bermuda triangle: players who’ve ended up there have disappeared into thin air.

While it seems fair to adjudicate both arguments—Thibs’ starters are playing too many minutes, and the reason is a lack of bench depth—as correct, we still know almost nothing about how the team’s current minutes debacle fits into a broader kahntext or about where it belongs when considering patterns of playing time in the NBA if the lens is widened to include other teams and other teams in past seasons. That’s what I want to look at in this post.

This post aims to do some basic comparative analysis to shine a light on where the current Timberwolves starting core’s minutes per game stacks up vis-a-vis other teams, both for the current season as of now, as well as for previous seasons going back to when Tom Thibodeau got his first head-coaching job for Chicago in 2010-11.

Specifically, I’m going to talk about (1) Thibs’ teams versus the average NBA starting unit mpg since 2010-11; (2) how the Wolves starting unit compares to other teams this season; and (3) for fun, whether any coach during Thibs’ tenure as an NBA head coach has ever relied more heavily on his starters than Thibs has so far this season.

Analytic Approach

To set up this analysis, here’s what I did: first, I got scraped all seasons of player data from basketball-reference starting in 2010-11—Tom Thibodeau’s first season as an NBA head coach—to the present (as of December 15th). To identify each team’s “starters,” I filtered the data to include only the top-five players per team and season in average minutes per game. So while I refer to “starters” below, for transparency’s sake, I simplified the concept to encompass a given team’s top-five players in average minutes played per game in a given season. I then took the average number of minutes played for each team and season from the MPG average of its top-five minutes grinders. In the course of the process, a few teams’ starting lineups, as depicted, here look different from the team’s regular starting lineup as you’d observe it while watching games, this approach to parsing the data enables us to see whether any team plays a core stable of players as relentlessly as Thibodeau does his current starting core.

What the Data Show

Thibs’ Teams MPG versus the NBA average, 2010/11-2017/18

Playing his starters high minutes is part and parcel of Thibs’ reputation as a coach. I wanted to see how this season’s team compared with previous Thibs-coached teams.

The table below breaks down the data.

Team Season MPG (Thibs) NBA avg. Minutes (Thibs-NBA avg.)
CHI 2011 32.8 31.9 0.9
CHI 2012 31.9 31.0 0.9
CHI 2013 32.6 31.6 1.1
CHI 2014 33.1 31.4 1.7
CHI 2015 32.6 30.8 1.7
MIN 2017 33.1 30.3 2.8
MIN 2018 35.5 30.5 5.0

In Chicago, Thibs’ starters typically played 1-2 minutes per game more than the league average for each season. During Thibs’ tenure in Chicago, his starters averaged 32.5 mpg. Since he took over Minnesota’s bench, his starters have averaged 34.3 minutes. In Minnesota, where his teams have had less quantity and quality so far—the narrative in ‘Sota, at least for this season, remains largely centered around making the playoffs—Thibs’ starters played almost three more minutes per game than the league average in 2017 and a whopping five mpg more than the league average so far in 2017-18.

This season is obviously the most significant outlier, and it’s the one we’re currently experiencing. It’ll be interesting to see whether POBO Thibs can add any depth to the current roster, which seems like the most probable way to bring the starting units’ minutes down in a #RegressionToTheMean kind of way.

Comparing the Wolves’ Starters to Other Teams in 2017/18

Unsurprisingly, Thibs’ Timberwolves starters lead the league in average minutes per game. As the table below shows, Thibs has played his starters an average 35.5 minutes. Jason Kidd’s Milwaukee Bucks starting unit—Giannis, Khris Middleton, Bledsoe, Snell, and Brogdon—is second at 33.44 mpg. Alvin Gentry’s New Orleans Pelicans (Holiday, Cousins, Ant Davis, E’Twaun Moore, and Rajon Rondo) rank third at just over 33 mpg. Rounding out the top-five are Doc Rivers’ Clippers starters (Griffin, Rivers, Gallo, DeAndre, Lou Williams) and Billy Donovans’s OKC Thunder (Westbrook, George, Melo, Adams, Roberson) at 32.8. [Editor’s Note: Recall that a handful of players like Lou Williams, who has averaged the fifth-most minutes per game this season for the Clippers get counted as “starters” as defined above.]

Team MPG Ranking
MIN 35.52 1
MIL 33.44 2
NOP 33.06 3
LAC 32.8 4
OKC 32.8 5
PHI 32.34 6
HOU 32.16 7
IND 32.1 8
GSW 31.84 9
POR 31.58 10
WAS 31.5 11
BOS 31.38 12
LAL 31.18 13
DEN 30.84 14
DET 30.76 15
ORL 30.62 16
MEM 30.44 17
CHI 30.38 18
DAL 30.22 19
NYK 30.12 20
MIA 30.1 21
CLE 29.92 22
UTA 29.35 23
CHA 29.3 24
ATL 28.3 25
SAS 27.96 26
BKN 27.55 27
PHX 27.52 28
TOR 27.44 29
SAC 25.46 30

How much do the coaches of the NBA’s best teams lean on their starters? The team currently atop the NBA, the Houston Rockets, are in the top quarter of the league in starters’ minutes per game. Moreover, Golden State’s Steve Kerr has not shied away from playing his starters significant minutes—the Warriors are 9th in the NBA at just under 32 mpg. The Spurs’ Gregg Popovich has stuck to his practice of keeping his players rested, as his Spurs starting lineups have averaged the fifth-lowest number of minutes at a hair under 28 mpg.

So while it is fairly obvious that there is a stronger correlation between talent and winning than there is between minutes and winning, the difference between running a starting unit 35.5 mpg (Wolves) and 32 mpg (Rockets, Warriors) adds up over an 82-game schedule and should be felt more later in the season and in the playoffs than it has been so far. And, as we know, so far the high number of minutes Thibodeau’s core rotation is playing has already attracted the ire of Thibs’ critics in wake of several disappointing fourth-quarter performances that led to losses.

Another way to skin the cat here is to look at whether high-minutes starting units tend to be good, bad, or mediocre. What do we know about the quality of the teams whose starters log the heaviest minutes? Looking at the top-5 from the table above, four of the five teams would make the playoffs if they started today.

  1. Timberwolves: 19-13, 4th in NBA Western Conference

  2. Pelicans: 15-16, 8th in NBA Western Conference

  3. Bucks: 16-13, 5th in NBA Eastern Conference

  4. Clippers: 12-18, 10th in NBA Western Conference

  5. Thunder: 16-15, 6th in NBA Western Conference

The Wolves currently have the best record (and would have the highest playoff seed). The others have been underwhelming: the revamped Oklahoma City Thunder, with its new big three of Westbrook, Paul George, and Melo, are one game over .500 and in a logjam with Portland and Denver for the fifth, six, and seventh spot in the West. The Pelicans currently have the eight-best record in the West, but aren’t even playing .500 basketball. The Bucks have MVP-candidate Giannis Antetokounmpo and are 16-13, good for 5th in the Eastern Conference. Each of these teams has one or two megastars, but is kind of thin otherwise. To be competitive, as all of these other than the Blake Griffin-less Clippers, have so far in their quest for a playoff birth, you get shorter rotations and more reliance on a small core. The best teams tend to have a couple of stars (Editor’s Note: Make that four in the case of Golden State.) but are also deeper and thus can afford to play their backups in longer rotations.

Has any coach ever relied more heavily on his starters than Thibs?

I thought it would be fun to see how Thibs’ current team’s workload compares to some of his contemporaries. The table below shows the top-20 highest mpg averages by starting unit since 2010-11.

Team Season Coach MIN Ranking
POR 2013 Terry Stotts 35.7 1
MIN 2018 Tom Thibodeau 35.52 2
POR 2011 Nate McMillan 35.06 3
LAL 2013 Mike Brown, Bernie Bickerstaff, Mike D’Antoni 34.88 4
POR 2014 Terry Stotts 34.74 5
MEM 2011 Lionel Hollins 34.5 6
GSW 2011 Keith Smart 34.42 7
BOS 2011 Doc Rivers 34.34 8
POR 2015 Terry Stotts 34.08 9
HOU 2014 Kevin McHale 33.92 10
WAS 2014 Randy Wittman 33.72 11
DEN 2011 George Karl 33.6 12
MIA 2011 Erik Spoelstra 33.46 13
MIL 2018 Jason Kidd 33.4 14
NYK 2013 Mike Woodson 33.24 15
GSW 2013 Mark Jackson 33.18 16
GSW 2014 Mark Jackson 33.16 17
NOP 2018 Alvin Gentry 33.12 18
CHI 2014 Tom Thibodeau 33.1 19
MIN 2017 Tom Thibodeau 33.1 20

Terry Stotts, ladies and gentlemen!

Portland Trailblazers coach Terry Stotts

(Editor’s Note: Patrick J makes a mental note to spend Christmas break in the Blazer’s Edge archives to see if the level of minutes hysteria in PDX in 2012-13 was as high as it has been in ‘Sota this season.)

Seriously, there are a lot of old-school names in the table above. Even the current coaches on that list, like Doc Rivers, now seem pretty old school.

Does that condemn Thibs’ program to the dustbin of history? Not at all. The Wolves have a ton of talent and are finally winning at a nice clip. But, that said, when you look at the table above, you don’t instantly think you’ve seen the future of NBA coaching before your eyes.

This post has tried to outline a few patterns to contextualize how Tom Thibodeau has used his core rotation this season. Although the minutes issue is unlikely to become uncontroversial during Thibs’ tenure in Minnesota, it should normalize—at least by Thibodeau standards—if and when Coach Thibs’ boss, POBO Thibs, does his job and acquires more depth.

Meanwhile, here’s a top-10 list of songs about being tired. Hopefully we won’t hear them blaring from the home locker room anytime soon.

Till next time.

Share this because Rubio would pass this along:
Tagged , , ,

4 thoughts on “Putting the Timberwolves Minutes Controversy in Context

  1. I’m not sure why this ‘debate’ won’t die. Obviously, it is unusual and almost certainly unwise, long term, to play starters this many minutes. However, Thibs isn’t likely to change this pattern suddenly. He tends to play his starters too many minutes. That’s a thing. His Bulls numbers aren’t as bad as one would think because he was often riding a few players too much not the whole unit. Still unwise, though. But as Patrick points out, this will normalize for him at some point (to closer to HIS mean). I think reasons it is so out or whack even for a Thibs team is because he desperately wants to (needs to for his job?) make the playoffs. He’s all in. Secondly, he kinda messed up in his creation of a bench and his spending, so he must feel an extra need to play starters. Of course this is his fault and there are in-season things he could do about it to fix it, but it is what it is at the moment. So on some level his leaning on the starters is understandable and in the short term it may work fairly well. But playoff teams need benches and the end of the season is going to be tough for us. Part of why things seem to be working so well for us (as opposed to last year) is an favorable schedule, a less strong NBA overall than expected and simply turning a young team into a vet team with 2 young guys on it. I mean, adding Gibson Teague and Butler automatically add about as many wins over last season as we have, and the rest can be explained by easy schedule, a little luck and playing our best players at an unsustainable clip. I don’t think Thibs’ coaching has fully been unclothed yet. It’s been wrapped in these temporary protections. To call him simply too old school (which is true) is to ignore the flexibility and talent issues inherent in the man. Late this season and in the playoffs he’ll have ample chance to prove me wrong.

    1. @pyrrol: Thanks as always for reading and commenting. We’re in agreement on a lot of things you mention above, i.e., there are still a lot of contingencies and unknowns to sort out. Concisely, it’s too early (still)–these factors will need to be accounted for in any credible/comprehensive assessment of where Thibs is going to be able to take the Wolves in the medium- to longer-term. I’ve gone easier on him so far this season than you have for this reason. But on a day-to-day basis, Wolves fans including me have had plenty of reason for concern and a lot to grumble about. Totally fair. I worry less about Thibs’ ability to coach a talented team with a little more depth–his heyday as Bulls coach was really impressive. Whether or not he can out-ace other teams’ GMs on personnel decisions, salary cap issues, and the draft remain to be seen.

      Off topic, but what do we realistically expect Justin Patton to become, when will it happen, and will Wolves fans ultimately be satisfied with him as our first-round pick at #16 last season? These questions are all part of a big mystery to me because I haven’t seen much of him. If he could actually add some depth as a rim-protector, even for just a handful of minutes a night, he could alleviate (put a band-aid on) at least some fraction of our depth issues.

      1. Haha, yeah, you and other writers here have a more gracious view of Thibs than I do. I think you guys have a more professional perspective on it, whereas I come from a pure fan place. As a fan, I have zero reason to like the guy other than maybe he brought in Butler and that made us into a winning team overnight. But that in itself has had less impact for me than I expected. After so many losing seasons, I thought I would be glued to every game and walking with a spring in my step just to have a winning season. But Thibs makes winning about the minimum fun it can possibly be, or at least that’s how it feels. So that makes me a little grumpy towards him. Beyond that, I’ve never actually seen him do anything with coaching and player development while here that has impressed me whatsoever. (And it is possible he’s overrated from his Chicago days where he had a favorable roster and was in the east when it was really bad.) It’s actually amazing, because almost every coach impresses me at some point, they tend to be impressive dudes. Taking most of the fun out of something that is supposed to be a fun distraction cuts down on brownie points real quick. A last funny bit–I don’t agree with Thibs on much, but I agreed with him getting rid of Dunn. It seemed like his perceived value was higher than his real value would ever be and that perceived value would crash soon. Turns out, so far he’s doing stuff in Chicago, so one of the few points of agreement I had with the guy was something we look to be totally wrong about. Of course, it is early, so we’ll see how it plays out (both Thibs’ all in season and Dunn). Beyond this season, the roster issues and keeping the parts we are ‘enjoying’ now is going to be a crazy puzzle.

        The thing with the Patton pick is that it was kind of a reach in some people’s eyes, yes, but also a redundant thing–we are a big heavy team with 3 centers. At least the pick we sold is a three point guy as well as a big. So, this decreases the likelihood of the pick being a rich value payback. However, if he’s a really good rim protector he could be very useful. The best near future use for Patton might be to allow us to trade Gorgui for wing depth or some other actual need. I like Dieng a lot, he’s fun to cheer for, but he’s kind of a tweener with PF and C aspects. His achillies on D is the same as Towns—he’s just not strong enough as a pure C, particularly in the legs. He doesn’t provide a gear change from KAT and he doesn’t play that well with him–Gibson and KAT are a better pair. Maybe the biggest thing that makes him expendable is that he has no standout skill. He’s a skilled big, but he’s just OK to good at everything. He has no calling card, no overarching skill that is a competitive advantage. Aside from lack of minutes in some games, this is the main reason Gorgui, a borderline starter on talent alone, doesn’t exactly dominate against 2nd string talent. We might be looking to develop Patton as a more athletic, rim protecting big we can replace Gorgui with and add depth. If things go great Patton could start and Gibson could be a big minutes bench guy. That’s of course crazy speculation and it is much more likely we don’t see Patton out of Iowa all season. Going forward, I see all this through a lens of not thinking KAT is best served as a C. Wouldn’t he be great next to a husky guy like Deandre? The problem with Patton is that, though the earliest of signs point to him perhaps being a better rim protector than KAT he doesn’t and might never have C strength, either. If this is the case but we want to go with those two, it might be nice to have a bench C who isn’t flashy but is very strong. Cole is basically a poooooor man’s version of KAT and G at the center position—weak. Off the top of my head, a guy like Kofus. Just a gear change we can play extra on nights were we need some beef. Well, that’s my hot take.

        Thanks to you guys for doing such a good job on the site!

  2. Stars in this league play a lot of minutes and most times their star counterparts do too. Milwaukee is playing The Freak and Middleton more combined minutes than Butler and Andrew are playing. An older player than either Butler or Andrew, LeBron James, plays more minutes the past decade than any other player and doesn’t seem to be getting more rest. Portland, OKC, NO and others have all had their best players playing heavy minutes. Factor that we have two young players that we want to play like stars and you now have 3/5ths of your starters playing the type of minutes that Superstars do. So it is really the other two starters that are playing more than most teams play their non-star starters. Taj and Teague. Let’s look at the reasons Thibs has for playing them heavy minutes.

    Taj is playing more than he should, partly because of the mystery man Belly and Patton being injured. I also think he offers the best low post defensive stopper that protects KAT and another veteran presence during key parts of the game. G has to give KAT and Taj breaks, so given that he hasn’t been as strong as in the past, you have Taj logging heavy minutes. Teague is the weird one. He hasn’t played overly well and Tyus has played better than expected, yet he still logs heavy minutes and Tyus gets maybe a minute more or less each game, but nothing significant. Crawford has been giving one or both of the wings a break now that he is shooting well and now we also have Marcus Hunt getting a little play. Thibs refusal to look small and have more guards and wings on his bench has limited his ability to play differently and give others more playing time.

    Thibs also plays slow and doesn’t have shooters that hit threes reliably. Because of that ,his point differential is naturally lower than running teams and so he doesn’t have many blowouts to clear the bench. All of this adds up to playing his starters a lot and his bench a little. If he had looked at the money he paid Teague and gave that to three players (Collison, Cunningham and Tony Allen leap to mind) he probably would play a deeper bench and could leave his starters on the bench. Now, I’m the first that says Thibs is out of touch with today’s offense and playing starters without developing a bench, but I think that even if Pop or Stevens had this team, they would play their three stars a lot and probably not go very deep on the bench, but they probably would have had the GM give them more depth and developed the players more.

Leave a Reply