NBA Basketball: The Spurs and Quantifying Movement

By Harrison Chase and and Carlos Peña-Lobel

As the postseason begins, the Spurs once again are in the driver’s seat of the Western Conference. This a familiar sight – since the 1999-2000 season the Spurs have amassed fifteen consecutive 50+ win seasons (including this year), the most in league history, and four 60+ win seasons. They’ve achieved this consistent brilliance through what Kirk Goldsberry describes as “clever tactics, unselfish execution, and ball movement” in a recent article for Grantland, which are the very things that AAU coaches all over the country love to emphasize. Announcers, fans, and analysts rave over the Spur’s superb unselfishness, but very little has been done to actually quantify how good their ball movement is. This has largely been due to the fact that no data has been available, but now with the release of SportVU data to the general public, there exists a way to quantify it. So, now that the data is available, just how well do the Spurs pass and move?

Let’s begin by looking at ball movement. The Spurs rank fourth in passes per possession, behind the Bulls, Jazz, and Bobcats. However, this doesn’t tell the whole story, because not all possessions are created equal. More specifically, not all possessions are the same length. The Bulls, Jazz and Bobcats all rank significantly higher than the Spurs in terms of possession length. This extra possession time naturally leads to more passes, as they have more time to do so (possession length and passes per possession have a correlation  of 0.5055, with an associated p-value of 0.002).

Therefore, in order to control for possession length, we plotted possession time against passes per possession and created a line of best fit. The line of best fits tells us on average, how many passes per possession a team with a given possession length would be expected to have. Using this line of best fit, we can calculate the difference between a team’s actual and estimated passes per possession. Standardizing these numbers, we can measure a team’s ball movement which we termed their “pass score.” We can do the same exact thing with movement, and create a “run score”. We would like to note that these metrics solely account for the relative quantity of passing and moving and don’t deal with the quality because there can be wasted passes and wasted movement. The pass scores and run scores of each team are listed below.

 Team Pass Score Rank Run Score Rank Atlanta Hawks 0.97 6 0.16 12 Boston Celtics 0.26 12 -0.45 18 Brooklyn Nets -0.28 21 -1.27 29 Charlotte Bobcats 1.55 2 1.54 3 Chicago Bulls 0.66 11 0.71 7 Cleveland Cavaliers 0.23 13 1.70 2 Dallas Mavericks 0.73 10 0.38 10 Denver Nuggets -1.24 26 0.03 15 Detroit Pistons -1.56 28 -1.12 27 Golden State Warriors -2.06 30 -0.52 22 Houston Rockets 0.04 17 -0.64 23 Indiana Pacers 0.17 15 -1.02 26 Los Angeles Clippers 1.09 3 -0.11 16 Los Angeles Lakers 0.80 9 0.07 14 Memphis Grizzlies 0.23 14 0.36 11 Miami Heat -0.18 19 -1.15 28 Milwaukee Bucks 0.98 5 0.39 9 Minnesota Timberwolves 0.07 16 -1.02 25 New Orleans Pelicans -1.42 27 0.93 5 New York Knicks -0.99 24 -1.96 30 Oklahoma City Thunder -1.12 25 -0.65 24 Orlando Magic -0.75 23 -0.35 17 Philadelphia 76ers 0.83 8 0.83 6 Phoenix Suns -0.72 22 -0.50 21 Portland Trail Blazers -0.12 18 1.32 4 Sacramento Kings -1.58 29 -0.49 20 San Antonio Spurs 1.70 1 2.60 1 Toronto Raptors -0.25 20 -0.49 19 Utah Jazz 0.90 7 0.66 8 Washington Wizards 1.05 4 0.08 13

Looking at the numbers, it becomes apparent that the Spurs are really, really, REALLY good at both ball movement and player movement. They’re slightly ahead of the Bobcats in passing, but blow all other teams out of the water in movement. Visually, the Spurs are a unique offensive team; this combination of passing and movement leads to plays like this. Few other teams in the league even strive for this, much less accomplish this so effortlessly, so often. Looking at a plot of these pass and run scores by team, we see this:

The red dot is San Antonio, all out by itself. The next closest dot to them is probably the Charlotte Bobcats who rank 2nd in passing, and 3rd in running. While this may surprise some, since they’re not considered an offensive juggernaut, it is worth noting that they have surpassed expectations this season, winning 43 games, way more than the 26.5 that Vegas set their over-under at.

As you can see, run and pass scores are seemingly correlated (p-value of .002) – teams that pass well usually run well, with a few exceptions. The green dots represent the three teams that have above average run scores while having below average pass scores. The green dot that has a barely above average run score is Denver, who have been known over the years for their up-tempo offense. Although they clearly took a step back this season, they still move fairly well. The green dot with a barely below average pass score is Portland, which has great running but could use more passing. The final green dot, all the way out by itself, is New Orleans. The Hornets have been an enigma this season, ranking top five in running but bottom five in passing. A possible explanation is that they are building a team around three dribble first guards. Poor Anthony Davis.

The yellow dots represent teams that pass well but lack in running. These teams are the Timberwolves, Clippers, Pacers, Rockets, and Celtics. A possible explanation is that players are spaced around the perimeter with one player driving and kicking, and then swinging the ball around the 3 point line for an open look. This lack of running but still above average passing is typical of teams who have guards that can penetrate (Rubio, CP3, Lance Stephenson, Jeremy Lin (Go Harvard!), James Harden, and Rondo) and then kick it out to their teammates who can nail threes.

We came up with a way to quantify ball movement and player movement, but do those things matter? While the Spurs have the best record this season, other teams with good pass and run scores have been terrible – Philly and Utah are both top ten in both categories, yet they are both among the league’s worst teams. Quantifying the effects of these things is the subject of our next post.

harvardsports

View all posts

• Amazing to think about. A little disappointed to see Utah and Philly in the top-10 of both, so I am looking forward to your next post. Keep on doing what you guys do.

• Bum says:

very informative!

• Mike says:

Very interesting. However, I find it difficult to believe this is what emphasized by most AAU teams. Not a ton of passing or off the ball movement in the typical AAU game. Or high school for that matter.

• Thank so lot excellent examples.and Quantifying Movement …great.

• Chris says:

For clarity purposes, pass score is the difference between expected and actual passes per possession and run score is the difference between expected and actual time per possession? So teams who have larger passes per possession pass better and teams with smaller time per possession values ‘run’ better?

How does this factor into their score value and rank?

Great stuff!