Nylon Calculus: Spacing Rating and why the Timberwolves may struggle

INDIANAPOLIS, IN - MARCH 28: Karl-Anthony Towns
INDIANAPOLIS, IN - MARCH 28: Karl-Anthony Towns /
facebooktwitterreddit

The Minnesota Timberwolves made seismic waves on draft night when the team traded for Jimmy Butler. Now, after signing Jeff Teague and Taj Gibson, it is becoming more and more clear that spacing may be tight with their new, rebuilt roster.

To evaluate just how much of an offensive challenge spacing will be for Minnesota next season (barring another drastic move), I put together a new metric for lineups, “Spacing Rating” and then looked at how the Timberwolves are likely to measure up next season.

Methodology

This model is more of a rough draft than a final thesis. Spacing is an abstract concept that can be difficult to fully quantify, so there are several assumptions that I have made (detailed below). I did my best in acknowledging and accounting for the factors that are key in this discussion and I hope to create an even more exhaustive model in the future.

As a dataset for comparison, I started off with every team’s two most-used lineups this past season. Then, I estimated each lineup’s 3-point percentage and 3-point attempt rate. To estimate 3-point percentage, I summed each player’s 3-point makes and attempts per 100 possessions to arrive at the team’s 3-point makes and attempts per 100 possessions. Then, I simply divided the 3-point makes by the 3-point attempts to arrive at the lineup’s 3-point percentage.

Next, I estimated the lineup’s 3-point attempt rate by summing each player’s field goal attempts per 100 possessions. I then divided the lineup’s 3-point attempts per 100 possessions by the lineup’s field goal attempts per 100 possessions in order to find the estimated 3-point attempt rate. Obviously, this isn’t accounting for any contextual factors with the lineup, diminishing returns or the ways the players in the lineup may interact with each other on the floor. However, since the spacing effect on a defense would seem to be more about a player’s reputation than their granular 3-point rate or percentage with a specific lineup, this seems like a reasonable approximation.

I estimated 3-point percentage (instead of using the on-court numbers from this season) because 3-point percentage is a particularly noisy statistic. Some five-man lineups did not shoot enough 3-pointers to truly explain the quality of 3-point shooting personnel that was present on the court. For example, a lineup of James Harden, Eric Gordon, Trevor Ariza, Ryan Anderson and Clint Capela would theoretically boast a lot of space, but it logged a 3-point percentage of just 32.9 percent this past season. In more minutes, this lineup would almost certainly inch closer to their expected 3-point percentage of 36.7 percent.

I chose to estimate 3-point attempt rate as well because some lineups (with just 200 or 300 minutes played) did not play enough minutes to fully exhibit the lineup’s offensive mindset and tendencies.

After estimating both 3-point attempt rate and 3-point percentage for each lineup, I translated these two categories into percentiles relative to all of the other lineups included in the data set.

I then averaged the two percentiles to combine the 3-point attempt rate and 3-point percentage into one inclusive number. Finally, I simply translated all of the averages into percentiles to arrive at the Spacing Rating. In simplest terms, Spacing Rating is the percentage of lineups that have less spacing than the lineup being tested.

Other factors

If you couldn’t tell from the explanation above, this is a pretty generalized estimation and there are several other contextual factors I did not include:

  • A shooter’s unique preferences catch-and-shoot or pull-up preferences and how those might be affected by the particular lineup
  • Variations in 3-point percentage from year-to-year
  • The spacing effects of mid-range shooting
  • Changes in shooting openness and frequency depending on the particular lineup
  • The impact of two traditional bigs on the court at the same time: With two traditional bigs on the court, opposing big men (players who are most likely to impact shots and clog the lane) are less likely to leave the paint. This could have an even bigger impact on court spacing than pairing a frontcourt non-shooter with a wing player who also struggles shooting the ball.
  • Coaching strategy: For example, the Hornets love to run shooters off a variety of off-ball screens to distract weakside defenders and open up the court, while the Cavaliers primarily utilize shooters in spot-up roles.
  • Player awareness: It is vital for offensive players to understand where to be and when to move relative to those around them. You can see how some players facilitate even better court spacing by moving in unison with a teammate’s drive in order to create better angles for a kickout pass.
  • “Outside-in” gravity: When we think about spacing, we mostly consider how an outside player impacts another player’s ability to get inside. The defense is forced to stick with lethal 3-point threats on the perimeter, meaning slashers have more room to operate in getting to the bucket. Interestingly enough, the mere threat of a player shooting a 3-pointer (their 3-point attempt rate) has a higher correlation with gravity than the player’s success from behind the arc (their 3-point percentage). In future models, weighting the 3-point attempt rate higher than the 3-point percentage makes sense.
  • “Inside-out” gravity: There is also an “inside-out” component to spacing. Players like DeAndre Jordan and Rudy Gobert are usually seen as a detriment to court spacing because of their lack of perimeter shooting, but these players can actually improve spacing by attracting the defense inside and consequently opening up opportunities on the outside.

Results

Here are the lineups in the top half of Spacing Rating:

And here are the bottom half of lineups in terms of Spacing Rating:

Next year’s Minnesota team, as many would presume, did not do well in this model. A lineup consisting of Jeff Teague, Andrew Wiggins, Jimmy Butler, Gorgui Dieng (or Taj Gibson) and Karl-Anthony Towns yielded a Spacing Rating of just 13.3 percent.

Neither Dieng or Gibson are 3-point shooters. Teague, Wiggins, Butler, and Towns all averaged less than 5.0 3-point attempts per 100 possessions (meaning they were all outside the top-200 in 3-point frequency among players who played at least 500 minutes last season). League average 3-point percentage last season was 35.8, and only Towns and Butler were above that mark, slightly. Teague and Wiggins were both below it, also slightly.

Next: Nylon Calculus -- A nearest neighbor approach to 2017 NBA Draft statistical comparisons

Of course, spacing isn’t the only facet to a team’s offense, and the Timberwolves’ shotmaking and primary play-making options will likely cover up a lot of their spacing concerns. However, this lack of space might ultimately limit the seemingly endless amount of potential in Minnesota.

Source: NBA.com/Stats