There’s a reason why a keyword search for “Oklahoma City Thunder” only calls up two posts on HoopChalk, with one of them being a lesson in how to defend Kevin Durant. Unlike the Miami Heat—who have redesigned their offense from the ground up to maximize the abilities of their superstars—or the San Antonio Spurs—whose sometimes vanilla-seeming playcalling masks myriad intricacies—the Thunder just don’t run that many plays.
Overall, their offense is second in the league at .99 points per possession, according to Synergy. The bulk of their plays are spot-ups (17.5%), although with Synergy’s way of categorizing plays, “spot-up” is more a term of exclusion than definition. If a pick-and-roll results in a kickout that leads to a pump fake, drive, and layup, that might be counted as a “spot-up.” Still, the fact that spot-ups form the bulk of the Thunder’s offense along with transitions, isolations (14.3% each) and ballhandler pick-and-rolls (13.3%) reveals a lot about how much it relies on individual playmaking and shooting.
The good news is that the things they do the things they do a lot very well. They rank 1st in efficiency in transition (scoring 1.21 points per possession) and 2nd in both isos (.9 ppp) and spot-ups (1.07 ppp).
Another thing the Thunder have going for them is consistency. Shifting to numbers from NBA.com, their starting lineup has, as of April 14, played 1307 minutes together—the most of any lineup in the league—and that lineup has posted the highest offensive rating (109.3) in the league. They’re not bad on defense either, posting the 6th highest defensive rating (97) for lineups with at least 400 minutes and their net rating (12.3) is second only to the Grizzlies’ starters with that same minutes qualification.
This is a team that gets good looks or gets to the line. They are 3rd in field goal percentage but 29th in field goals attempted while being 2nd in free throws attempted and first in free throw percentage. Given their high shooting percentage, it’s maybe not surprising that they’re 25th in offensive rebounding (the Heat and Spurs are both worse at offensive rebounding). They also take care of the ball (28th in turnovers) and protect the rim (1st in blocks). The bottom line is that they put the ball in the hoop and keep the other team from doing the same, even though they don’t employ a lot of ball movement (21st in assists).
Taking all these numbers together and from watching tape, I can liken their offense to two fairly different things: the principles of basketball video game offense and the John Coltrane quartet.
Here’s the thing about playing on offense in basketball video games, as a rule: The guy with the ball is always the best player on the court. This is because the guy with the ball is controlled by a living, breathing human being and not just a set of tendencies. For this reason, you’ll find it’s difficult to run nuanced offenses like the triangle or the corner. While you control the guy with the ball at the elbow, the computer is running your other guys in a ham-fisted fashion around screens. A lot of the time, your other players will pop out too far or not far enough or will clog up the floor in weird spots. If you try and play off the ball, you have to rely on the computer’s very rudimentary decision making based on patterns and tendencies.
So what’s your best option? Simple pick and rolls that clear out the floor and let you work (a good example of this in NBA 2K13 is Red Angle Iso from the Bulls playbook, which I wrote about for Operation Sports). Here are Russell Westbrook and Kevin Durant doing just that.
Everyone else clears out to let Durant set the screen for Westbrook. This is a real damned-if-you-do-damned-if-you-don’t set because Durant is a huge shooting threat (45.3% from midrange) and also a terrifying driving threat (70.4% shooting in the restricted area), while Westbrook might be a slightly less reliable shooter, but is at least as good a finisher (41.3% of his points come in the paint).
Another good option are floppy sets that spring shooters. Here we see Durant pop out from a half-screen by Serge Ibaka and then set up above a screen from Kendrick Perkins.
Technically speaking, I’m not even sure it’s a true floppy, but the basic principle of having a shooter pop out from behind staggered screens is the same. This is another staple of video game offense because one of the easiest things for the AI to do is put big bodies in the way of moving guys.
Once one of the Thunder’s two best guys get the ball in space, they either pull up and shoot, back their man down or try to drive the lane. They either get the bucket, draw the foul, or kick it out to a player like Ibaka, who has a solid midrange jumper, or players like Thabo Sefolosha or Kevin Martin, who can generally be trusted to knock down open 3-pointers.
Then there’s the signature pin-down screen for Durant, which Sebastian Pruiti detailed far better than I could hope to over on Grantland. Whichever of these plays the Thunder are using, the goal is simple: get one of their best players the ball with space to shoot, drive or create, or get it to one of their other players where they just have to make one correct decision about shooting. That’s video game basketball.
For a more nuanced, philosophical approach, though, let’s turn now to the John Coltrane quartet.
A lot of what gives jazz its shape and direction is how the improvisers relate to the underlying structure of the song they’re playing. Without getting too much into the technicalities of it, bebop approaches each chord change as its own tonal center. Extensions are piled on top of chord tones and then the changes are executed at blinding speed and with precision; the magic comes from how the player stitches together the changes with a melody that addresses and encompasses them in an exciting way.
I would liken the Spurs to bebop. Each piece of each playset—from screen to roll to flex screen to handoff to re-screen and pop—is manipulated and exploited, threaded through with an understanding of where to break it off and where to follow through. A player like Parker or Duncan has become so good at this system that what is actually tremendously difficult to do can look easy and effortless.
When Miles Davis recorded Kind of Blue in March of 1959, it introduced modal jazz, which didn’t pile chords on top of chords, but instead located the players within a general tonal center that they were free to do with as they please. John Coltrane was in the sextet that made the album, and the quartet he put together with drummer Elvin Jones, pianist McCoy Tyner and first bassist Reggie Workman in 1961 and then Jimmy Garrison in 1962 would build on this concept with a flexible but clear hierarchy among the players.
The tunes were not found structures to be navigated parkour-style, mazes to be solved. Instead, standards like “Summertime” or “My Favorite Things” and originals like “Equinox” and “Chasin’ the Trane” are open fields where the players can build and raze ideas over and over again. The limitation is not in the song, but in the player’s ability to make something with it.
In this formulation, Durant is Coltrane—prodigious, titanic, a creative force making things happen from outside and inside with a nearly unbelievable combination of natural talent, vision and tireless dedication to getting better. Westbrook is Jones. The ultimate foil to Coltrane, Jones was a propulsive, percussive force of nature, often turning the drums from timekeeper to lead soloist. Jones roils the kit, churning the songs’ into an unstoppable maelstrom. Sefolosha and Ibaka (and Martin) play the role of Tyner. Rarely on the frontline of the improvisation in the quartet, he instead played the more measured counterpoint to the fury and virtuosity of Coltrane and Jones. (In some ways, Harden is an even better comparison, since Tyner has had his own very successful career as a leader.) Perkins is the oft-overlooked Workman and Garrison. No one expected them to model and reshape the tune the way the other players in the group did, but they provided stability, forged a solid base for others to take flight from.
When things are clicking for Oklahoma City, the offense should look very much like a Coltrane performance: there’s a kind of inherent unbalance to the way the unit is structured, but so long as the individual talents of the players are marshaled well and don’t fail them, they can create beautiful things out of relatively simple material.
The problem for the Thunder is likely to come when those individual talents fail them. On the one hand, all those minutes together have made the starters very comfortable with each other, but on the other, that’s a lot of minutes for a team to play. A lot of their success last season—especially against the Spurs—came from being able to run their opponents out of the gym. Certainly the compression of the schedule had its own deleterious effect on teams’ health, but the Thunder’s starting lineup played just a little over half as many minutes last year (664) during the regular season as they have this year. As good as they are, and as much better as they’ve gotten this season, their success is still grounded in individual virtuosity. It remains to be seen how they’ll deal with opposing teams that can find ways to consistently derail their overflowing, video game-esque offense.