There were many reasons why the Thunder lost Game 1. Examples include being out-rebounded by 15, Patrick Beverley being the bane of our existence, Victor Oladipo’s non-existence, and more.
But the biggest problem was the pick-and-roll defense. It was atrocious. Don’t just take my word for it, here was what Steven Adams had to say after the game:
“The bigs, especially me, played trash on the reads, and the pick-and-rolls were absolute garbage. We have to get back to the drawing board and kind of look at how we can give a better effort in Game 2.”
What exactly was so bad? I went back and re-watched the game to find out, charting every pick-and-roll situation in the first three quarters (the fourth quarter was essentially garbage time). Let’s start with the most general findings and then zero in on where the real problems occurred.
Houston ran 70 pick-and-roll situations through the first three quarters.
The pick-and-roll is the basis of any NBA offense, but the Rockets take it to another level, running PnR action on essentially every half court possession. On the Fox Sports Oklahoma broadcast, Michael Cage cited their average of 64 on-ball PnR’s per game. The 70 I counted through three quarters is unofficial and includes some double actions that will be discussed later.
James Harden was the ball handler in 54 situations.
Again unsurprisingly, Harden dominates the ball and receives pick-and-rolls at an incredible rate. This number is compared to just 36 times where Russell Westbrook was the PnR ball handler over the three quarters. Interestingly, when Harden leaves the game the Rockets continue to run a high rate of pick-and-rolls, whereas the Thunder’s offensive focus switches to post isolation with Westbrook off the floor.
Harden’s PnR situations led to 42 points.
Of Houston’s 89 total points during my sampling, 47 came as a direct result of half-court PnR’s (52.8%), which I defined as an immediate score or assist, a hockey assist, or an isolation score as a result of a switched PnR. Of those, Harden’s PnR’s accounted for 42 of those points (17 individually scored, 6 assists/other).
In practical terms, nearly all of them were directly related to Harden. So let’s hone in on those.
Of those 42 points, 30 came from the Thunder switching or sagging.
How should the Thunder handle a Harden PnR? Well they started out the game fully switching nearly every screen, a strategy they used to great effect during last year’s playoffs. It didn’t work this time around. The defense changed, either by purposeful adjustment or the big men simply getting fatigued, into a “sag and recover” approach designed to allow Harden room to shoot from the outside where he had a low percentage on the game.
Conventional wisdom tells you that if a player is quicker, you’re supposed to back off and give them space. But any good basketball coach will tell you this is the exact opposite of what you should do. In order to guard freak athletes, high level players are taught to crowd their man, not giving him any room to gain balance or build momentum. The “sag and recover” strategy merely allowed Harden to turn downhill, coming full-speed ahead at the Thunder bigs. They had no chance, giving up 13 points on 15 situations in which this was utilized.
Kanter got the worst of the treatment. In the six PnR situations in which he was involved, he sagged on four and switched on two. Houston cashed in for eight points, leading Billy Donovan to say this:
P&R lob to Capela causes Billy Donovan to tell Mo Cheeks, "Can't play Kanter." pic.twitter.com/4ROFSNpFw6
— Yaya Dubin (@JADubin5) April 17, 2017
9 “combination screens” were used to get Harden favorable ISOs.
The Rockets were smart about how they approached the switching, using “combination screens.” Knowing that the Thunder weren’t likely to switch a screen involving a guard-center combo like Roberson and Adams, Houston would first send up Ryan Anderson to set a pick. The Thunder were willing to switch Anderson’s man, either Taj Gibson or Jerami Grant, onto Harden for a guard-forward swap and trust that they could stay in front.
Then Harden would simply back the ball out and wait for the screen they really wanted, coming from center Clint Capela, bringing along Adams or Enes Kanter, and the Thunder were willing to switch forward-center Gibson/Grant with Adams/Kanter. In the span of five seconds, Harden had an isolation situation against a Thunder center. If you watched the game, you know the results.
— NBA (@NBA) April 17, 2017
On the other end, Ryan Anderson was put in just 10 PnR’s.
The most alarming discovery was the Thunder’s inability to play the same mismatch game on the offensive side of the floor. While the Rockets were continually searching for mismatches for Harden, the Thunder looked as if it had the same game plan as any Tuesday night regular season game in January. The Rockets put Adams and Kanter, their two prime targets, in 34 PnR’s to just 17 PnR situations in which the Thunder targeted Ryan Anderson and Nenê, the weakest links in the Houston defense.
Of those 17 situations, the Rockets made a conscious effort to hedge all of the screens, showing their big man for a second before the original defender hustled through the screen and back to Westbrook. Which meant no mismatched isolation possessions for Russ, and no fancy NBA slow-mo videos of him shattering Anderson’s ankles.
How do we fix it?
The problem once again stems from three-point shooting. The Rockets have an abundance of it, and with the floor nice and spread it makes it nearly impossible for the Thunder employ the hedging strategy. On the other hand, OKC’s inability to space the floor allows for plenty of help on Westbrook’s PnR drives.
On the offensive side, it’s pretty simple. Oladipo needs to start hitting from deep, and Roberson needs to remain in his dream-like trance. From there, coach Billy Donovan will need to decide if more minutes for Alex Abrines and Doug McDermott are required to space the floor, even if it means some sacrifice on the defensive end.
In order to stop the onslaught of the Harden PnR, perhaps the Thunder should consider double-teaming and trapping him? While it’s risky and could result in Houston getting open looks from behind the arc, it will at the very least get the ball out of Harden’s hands and make other players beat them.
Luckily, Billy Donovan is a lot smarter than I am, and will have thought about other possible solutions before Wednesday night’s contest.
How will the Thunder handle Houston’s pick-and-rolls in Game 2? I can’t wait to find out.