It’s reasonable, if not guaranteed, to expect that the Toronto Raptors and Houston Rockets won’t play each other again this season. They’d be a hipster NBA Finals matchup — a cute pick but one that’s way too far away from happening. Let’s not expect that.
How the Raptors’ defense gave the Rockets fits, and how Houston responded
This is the defensive scheme Toronto used to force Houston into their worst offensive half all year.


But Friday’s highly fun 108-105 Raptors’ victory could still have postseason ramifications thanks to the first half. It was then Toronto bullied their opponent into a 16-point opening quarter, the lowest Houston had recorded all season. For 12 minutes, Houston’s offense — second-best in the league — crawled to an uncomfortable mess. How’d that happen?
Not everything the Raptors caused on Friday is replicable without their personnel, of course. Not everyone has a guard like Fred VanVleet, who can pressure opposing guards full-court — even Hall of Fame ones like Chris Paul — and force him into turnovers like this play right here.
The Raptors also emphasized the transition defense, and they held Houston to just five fast-break points and nine three-pointers by halftime. Again, easier said that done. Still, there were actual schematics behind those numbers.
Here’s where Toronto really screwed with the Rockets.
The Raptors’ pick-and-roll coverage was the biggest reason for Houston’s season-low in the first half. With two massive big men, Jonas Valanciunas and Jakob Poeltl, Toronto opted for an aggressive ICE coverage, or something even reminiscent of the old Roy Hibbert days when he and the Pacers smothered teams. Here were the rules:
- Toronto’s big man stayed far back in the paint, rarely coming out past 10 feet, on pick-and-rolls. This set up a constant presence under the rim for any driving guard. By not switching, this prevented easy Harden or Paul isolations that usually makes up Houston’s offense.
- The Raptors did switch on non-big pick-and-rolls, trusting all their perimeter players could adequately match Harden one-on-one. That’s the beauty of Toronto’s personnel, which has few weak links, even if Houston’s star did eventually end up with 40 points.
- The defending guard in the pick-and-roll went over screens aggressively and crashed back into the play. With poor defenders, this might not work, but Toronto made it count.
- Toronto’s help defenders had one primary goal: prevent three-point shots. Against Houston’s weaker shooters, Luc Richard Mbah a Moute and P.J. Tucker, there was some room to sag into the paint and recover. Against Trevor Ariza or Eric Gordon, Toronto’s perimeter defenders stuck to them doggedly.
Here’s what it looked like:
Or this:
Sometimes, Harden would get Toronto’s paint player to bite on his drive, enough to open up a lob for Capela. Three of Harden’s four assists happened on plays exactly like this.
But the two Houston guards combined for just seven assists — this Toronto defense effectively took playmaking out of their hands. Often, the only available pass to Capela was one that asked him to make plays five or more feet away from the rim. Make no mistake: Capela’s 6-of-11 shooting night (3-of-8 without the lobs) and his three turnovers is exactly what Toronto wanted. The 23-year-old Swiss center is an excellent player, but he’s not a superb decision maker or clever in space.
Another smart layer to Toronto’s coverage: When their perimeter defense was stuck on Harden or Paul in isolation, they overplayed their shooting hand and invited the drive. The entire gameplan revolved around making Houston challenge Valanciunas and Poeltl at the rim, and making sure those two weren’t wasting energy chasing people around on the perimeter. For one half, it worked.
Here’s how the Rockets adjusted.
Houston scored 62 points in the second half, nearly coming back to win despite their first-half struggles. Much of that can be credited to Harden, who was meticulous with his drives to the rim and nearly impossible to stop, even if Toronto’s defense was set up exactly like they wanted. Look how many times Harden takes on Valanciunas or Poeltl sitting right under the rim and scores anyway, or tosses in a floater even with a defender on his hip.
With Capela on the court, Houston stopped forcing him into Harden pick-and-rolls and instead tucked him out of the way on the baseline. Ariza and Tucker both tried running pick-and-rolls for Harden, with varying results. Since neither player is a natural pick-and-pop player and Toronto could just switch, these possessions often devolved into difficult Harden isolations. Sometimes, a non-star Rocket would be asked to create late in the clock, which is always a win for opponents.
Paul actually used Capela pick-and-rolls to aggressively free himself for a couple mid-range looks, something Houston surely would’ve used more if Harden hadn’t got going. Paul’s a historically good mid-range shooter, enough that Houston values him taking jumpers inside the arc, something they typically discourage. Notice Capela’s pick here, set solely to free Paul with no thought of rolling.
That said, Toronto’s defense forced Houston into small ball. Capela only played 11 minutes in the second half, and the Rockets never looked at Nene again after a disastrous four-minute stretch in the first quarter. The small ball spaced out Houston’s offense once again but gave up more on the other end, as it often does. It was enough for Toronto to pull out the win.
Ryan Anderson’s absence with a hip injury was notable. With his natural pick-and-pop game, it would have been impossible for Toronto to sink their big men so deep. If the Raptors switched that pick-and-roll, it would’ve given favorable isolations to Harden that Houston would have exploited all game. Anderson may play a smaller role in the postseason, but his worth still shows up clearly in situations like this, when Houston needs a coverage-busting shooter.
Toronto’s approach was smart, and it still really only held up for a 24 minutes. But it at least made Houston uncomfortable, and that’s something few teams can say. Rockets opponents in April and May will look carefully comb through the results of this first half, and see if there are any ideas to carry with them.















