Using SportVU Motion Data to Examine Jazz Roles

NBASavant.com

SportVU motion tracking data has brought a vast wealth of new information to basketball nerds everywhere. Even while only utilizing public data that’s barely the tip of the iceberg of what’s available to teams and high-paying clients, the ability to track the movements and actions of all 10 players on the court plus the ball at the granular level has broadened our potential understanding of the game exponentially.

One area mined less frequently than others, mostly due to lack of public availability until recently, is simple player movement. We’ve had the ability to look up elements like top speed and total distance covered since SportVU data became available on NBA.com a couple years ago, but the actual areas of the court these movements covered weren’t available.

That changed with the introduction of movement charts last year, which digitally track every play recorded by all 30 teams. In theory, this advance could reveal a huge percentage of SportVU’s total data set — nearly every second of NBA action can be found in these motion logs. In practice this has been much more difficult; the amount of raw data involved is ridiculously massive for your average blogger, and even beyond concerns like data storage, the ability to correctly parse such huge wells of information into useful sub-categories is challenging for most in the public sphere.

Luckily, we still have a few benevolent geniuses among us, not the least of whom is NBASavant¹ curator Daren Willman. Daren’s sites have been at the forefront of publicly available tracking data for some time, and he’s recently begun making forays into the usefulness of NBA player movement. An example:

NBASavant.com

What you’re looking at is a hexagonal heat map showing the most common areas Kyle Korver inhabited (offense on the left, defense on the right) during Atlanta’s game against Washington Saturday night. Unsurprisingly, he spends a lot of his time offensively outside the 3-point line, though as one of the more active shooters in the league who utilizes plenty of off-ball action to find openings, he spent a reasonable amount of time inside the arc as well.

Though this data is yet to be available publicly, Daren has kindly shared a sliver with me. Let’s break down these heat maps for several relevant Jazz players and see what the data teaches us, or simply confirms what we already knew. We’ll primarily use Utah’s win over Denver from last week as our barometer.

Bigs

Gobert:

NBASavant.com

Favors:

NBASavant.com

Booker:

NBASavant.com

Gobert’s chart probably contains the fewest surprises of any Jazz player. As a volume interior player who almost exclusively defends others of the same ilk, Rudy has very little reason to ever stray from the middle third of the court horizontally. His map from near midcourt highlights just how little variation there is — he’s virtually always running up or down the court on a line directly parallel to both baskets. The only exceptions here will be transition chances, but these are few and far between for Gobert, who’s much more often the guy initiating the break with an outlet pass.

Favors and Booker offer a bit more variation, and a few points of interest. Derrick is the Jazz big whose charts vary the most from game to game — this chart against the Nuggets came on only 15 minutes as he battled flu-like symptoms, and showed a bit more frequent instances of Favors operating outside the paint than many of his other games. For instance, look at his chart against Memphis:

NBASavant.com

Not surprisingly, spending much of his time against guys who operate down low in Marc Gasol and Zach Randolph, Favors was inside the paint more often on defense. Offensively, it’s clear the Jazz didn’t want him on the block quite as often, particularly when Gasol was guarding him — Favors attempted just three post possessions on the game after averaging over five a game (including his shortened night in Denver) up to that point, per Synergy Sports. Two were on Randolph, and one was a late-shot-clock dump-in against Gasol. It’s clear that of Utah’s starting bigs, Favors’ role will be the one that varies most often depending on matchups.

Booker also has a couple points of interest, and much of the data perhaps indicates why the Jazz’s offense mostly stagnates with him² despite his several above-average skills on this end. His chart above is pretty similar to this one against Portland:

NBASavant.com

Notice how much time Trevor spends in the midrange areas offensively? This isn’t much like Favors, who spends a lot of his time in those areas either posting up or receiving short roll passes in pick-and-roll and shooting. In Booker’s case, these are most commonly his spot-up locations — why not back up a few feet? I’ve loudly wondered recently whether an uptick in attempted 3s from Book might open things up for Utah’s entire offense while he’s on the court, but these heat maps seem to indicate that his preference, and perhaps Quin Snyder’s preference as well, is for him to remain inside the arc. One can’t help but wonder if things would look a bit different for Utah’s second units if he could comfortably space out deep.

Wings

Hood:

NBASavant.com

Hayward:

NBASavant.com

Burks:

NBASavant.com

Each of Utah’s three primary ball-handlers on the wing has a somewhat similar chart, though a look through each individual contest reveals that their higher-volume areas from beyond the arc tend to vary from game to game. Generally, though, each of the three is fairly well spread-out offensively — they’re all guys who initiate plays up top but are more than capable of penetrating, and therefore spend time virtually everywhere in halfcourt offense. It’s much of the same defensively, as they’ll typically guard similar players.

Burks’ chart against Denver has one point of interest. Compare it above to his map against Indiana a few days prior:

NBASavant.com

Notice how much less time he spent inside the arc, particularly close to the basket, against Denver? It’s not a coincidence that this Nuggets game was one where Alec clearly had the coaching staff in his ear about opening the game up for his teammates, evidenced by his eight assists that night. His heat map helps showcase this — so much less time spent in the paint indicates that he was cutting many of his drives short to find teammates elsewhere.

Ingles is a much different case:

NBASavant.com

Joe spends almost no time inside the perimeter offensively — and this Nuggets game was one where he actually ventured there even more often than usual. Check out his split from the previous night against the Blazers:

NBASavant.com

This is a spot-up shooter in the purest form. Shoot, even Korver spends significantly more time in the paint. Ingles hasn’t been utilized as the sort of off-ball rocket Korver is, and almost exclusively functions offensively as a spacing agent capable of putting the ball on the floor only if absolutely necessary. To be clear, this isn’t a bad thing — at well over 40 percent from deep dating back to the All-Star break last year, Ingles absolutely should be operating outside the arc as often as possible to help goose Utah’s spacing.

Point Guards

Burke:

NBASavant.com

Neto:

NBASavant.com

The largest difference between Utah’s two point guards, despite his shooting concerns coming over to the NBA, has been the chunks of time Neto spends as nothing but a spot-up shooter on the perimeter. His big hexagons there from the Denver game were actually on the smaller end of what we’ve seen for the entire year — he’s utilized as a pick-and-roll ball-handler from time to time, but the Jazz seem to go away from this when they really need baskets.

Burke’s own chart varies from game to game pretty heavily, and this isn’t a surprise. There will be few point guards in the league who are primary offensive initiators and don’t touch a pretty large portion of the court offensively; creation often takes a guy all over the floor, especially guys like Trey who are unafraid of keeping their dribble going as they probe around. He’s spent time as an off-ball spot-up threat, though not as much as Neto, and both have mostly similar defensive heat maps as one would expect.