National Basketball Association

2016-17 NBA Preview: Minnesota Timberwolves

Published Jun. 30, 2017 6:28 p.m. ET

foxsports

The Minnesota Timberwolves are one of the most promising teams in the league, and they also create some of the most divisive opinions about their players. Now with a respected coach coming in and their two returning number one picks, people are seeing this propitious roster delivering on its potential in the present, not the future. This has caused a chasm between the projected win totals from some analytics systems and the general public. But there are a few factors that we can quantify to explain some of the difference, as well as a few reasons to be cautious about the team hitting 0.500 or better.

“It takes a very long time to become young.” – Pablo Picasso

2015-16 in review

With the exception of Karl-Anthony Towns and his marvelous rookie year, Minnesota had a terrible season. They won 29 games with one of the worst defenses in the league, and Andrew Wiggins, who they paid dearly for, only made strides with his scoring — he doesn’t yet have the type of all-around game you’d expect from a superstar. They had once again one of the lowest proportions of three-point shots in the league, and they made a low percentage when they actually attempted them. They did, however, have the highest rate of free throws in the league, thanks to their guards Andrew Wiggins, Kevin Martin, and Ricky Rubio. But few people voiced their concerns about the team because of Towns’ historic season; their minds were distracted by what he could become.

Rotation players in: Kris Dunn, Cole Aldrich.

Rotation players out: Kevin Martin, Tayshaun Prince, Kevin Garnett

The Timberwolves had an odd but maybe effective strategy: sign veterans who can help the locker room and act as mentors to the young kids, but make sure they’re so far past their prime they can’t possibly affect your chances at striking it rich in the lottery. Martin and Prince were beautiful fits in that respect, and Garnett too for that matter. Martin left midseason and Prince was not signed over he summer while Kevin Garnett yelled his way off into the sunset.

Kris Dunn was loved across a wide swath of translated stats, like Nick’s stats and 538’s CARMELO system, which actually sees him as an above average player as a rookie — that’s rare. He’s a big point guard with a killer wingspan, and a possible future replacement for Ricky Rubio. Cole Aldrich was loved by the numbers too, as his per minute defensive stats were some of the best for any bench player in the league. Similar centers were paid about twice as much on the market, and with their glut of big men on the team it’s unfair Minnesota got such a great deal.

2016-17 projected

More from Nylon Calculus

Many projection systems use some combination of an adjusted plus-minus metric and box-score plus-minus, and from that starting point the Minnesota Timberwovles are a bewildering team to predict given their betting lines. For starters, two of their core players, Andrew Wiggins and Zach LaVine, look pretty awful by advanced stats, and their minutes should be pretty heavy this season. Zach has some of the worst adjusted plus-minus stats in the entire NBA, but for most of his career he’s been playing out of position at point guard. With the rookie Kris Dunn and the change during the last half of the season where Zach was playing off-guard more frequently, that will no longer be a major issue. Zach LaVine’s DRE, basic statistical plus-minus metric where 0 is average and 5 or greater is superstar level, increased by 1.5 points at shooting guard. It’s an issue most projection systems can’t fix automatically.

Then there’s the issue of Andrew Wiggins. The famed prediction guru, Nate Silver, uses a player projection system called CARMELO, which finds similar players based on a variety of stats for an estimate of a player’s growth. Since his most similar peers improved considerably in the corresponding season of interest, they have his rating skyrocketing by nearly two points per 100 possessions — and yes, for projection systems, which are inherently conservative, two points is a lot. While I don’t see quite the same jump in my numbers, he should improve by a significant amount. Conversely, since Karl-Anthony Towns was so great in his rookie season, I see him basically playing at the same level in his second season.

Somehow, a large chunk of Minnesota’s improvement comes from a rookie, who looks decent even in his first season, and a cheap center in Cole Aldrich. They’re surprisingly deep now, at least in the frontcourt, and how good they could become will depend on their rotation and how they dole out the minutes. Zach LaVine, as discussed earlier, has some pretty ugly advanced stats, so the numbers say they could improve simply by slicing down his minutes. They should also minimize Jordan Hill’s minutes and replace them with Aldrich’s or Dieng’s. And of course, Ricky Rubio’s minutes are crucial because he scores well in about every modern metric. He’s another reason why their win projection is so volatile because his injury history is pretty deep for a guy his age — not to mention the persistent rumblings that he may eventually be traded.

Even with a careful alignment of minutes and a kind projection for Andrew Wiggins and Zach LaVine, the Wolves still fall short of the 0.500 win percentage goal. There’s yet one factor not accounted for, and it’s pretty important: they’re bringing in defensive superhero Tom Thibodeau. It’s hard for anyone to deny how he’ll have a positive impact on the team, but coaches are notoriously hard to quantify. For example, he’s had a lot of success in the past, but how exactly does that explain the future? Coaches are great until they simply are not anymore. Perhaps the league has learned most of their tricks — or perhaps they run into bad luck. But even conservatively speaking he should have some sort of boost. Most analytics predictions ignore coaches, and in this case it appears to be a mistake.

Quick statistic

The future of Andrew Wiggins is one of the most controversial topics in the league, but if he doesn’t improve in his third season the number of critics will grow. While FiveThirtyEight is optimistic about his upcoming season, I don’t think his “most comparable” list is all that encouraging. Using my own stats, which only go back to 1997 but I can consider non-box score numbers, his most comparable list is sobering, except for a young Kobe Bryant popping up. As you can see below, DeMar DeRozan is the name that comes up most frequently. Wiggins has some pretty mediocre peripheral stats, but many are important, like his steal rate. He looks great as a scorer for his age, so he’s getting compared to one-dimensional scorers. I know his defense is expected to improve dramatically, but through two full seasons he’s been lacking in key areas that don’t often improve with age.

Summary

The young pups in Minnesota look like an inevitable playoff team, but there’s a disagreement over whether or not it’ll be as soon as this season. The team has some pretty extreme numbers for a mediocre team, as Ricky Rubio and Karl-Anthony Towns look like stars while Andrew Wiggins and a handful of others look like train-wrecks or replacement level at best. With a careful algorithm, those problems can be mitigated but not entirely. Finally, the biggest reason for an improvement is Tom Thibodeau, who’s inheriting a team that nearly ranked last in defensive efficiency. It’s hard to imagine a team of his performing like that, but it’s also hard to quantify how valuable he will be.

Win predictions:

Mine: 37.6. A blend of several metrics, including Dredge, with a few other factors considered, like coaching.

Andrew Johnson’s: 30. A combination of PT-PM (a SportVU player tracking metric) blended with RAPM. Two-time reigning champion of the APBRmetrics board predictions contest.

Nick Restifo’s: 32. A simulation using BPM and RPM for player value, which includes game effects like altitude and rest.

Kevin Ferrigan’s: 35. A player projection system with inputs from RAPM, BPM, height, and age.