So, I've been curious for a while now. Once in a while, somewhere there will be a mention of the Rockets defense, and some generic statement of how well it's doing during some measure of time. The offense has been explained many times and in a lot of ways; those explanations makes sense, but I haven't really understood what's making the defense tick. At a cursory glance, it's not really obvious. Watch a Rockets game and you'll see a lot of active hands. But is the 1on1 defense that great? Not really. Are shots contested more? Is Clint really more so of an interior anchor compared to other centers? Hmm. What about wing defense? Most teams nowadays have their versions of 3&D - are the Rockets really better defenders on the outside? So I thought I'd do some digging myself. First of all, let's try and clear up any potential misconceptions: The Houston Rockets defense is good, but it's not amazeballs. Most overall defense metrics will have them somewhere about a third of the way down at around number 10 in the NBA. That's still really good and a huge improvement from last year where they ended were closer to 20th. More importantly, it's enough for their offense to take over games. The team is actually really bad at contesting shots, near the bottom of the league. So whatever they're doing, it's not because they're up in their opponents' grill when they shoot it. Opponents actually shoot pretty well. The Rockets don't just contest shots, they don't contest 3's either. Ok, so it's not about contesting the more valued 3 ball either. The interior defense is also near bottom of the barrel, but slightly better than last year. I guess I see this in two ways. One, yes they aren't lockdown defenders in the paint, traditionally the highest FG% of all shots. But two, you have to remember the Rockets shoot more 3's than anyone else. Maybe the philosophy of this part of the defense is, "Come on in and shoot 2's. We'll look like we're giving it to you more than 3's, even if we don't contest that either. Eventually we'll outscore you." Is this a reach? I don't know. What the Rockets do well: Getting Steals, 2nd in the league per 100 possessions. Easiest way to lower the opponent's PPP? Don't let them shoot it in the first place. This really starts with the acquisition of Chris Paul at a team leading 2.8 steals per 100 possessions. But you keep looking down the list - Harden has stepped up at 2nd and Ariza not far behind at 3rd. And on top of this, Luc Mbah a Moute, PJ Tucker, and Nene have all been doing their part. As an extension of the above, deflections and guarding the passing lane. This is probably the most visible part of their defense that is explained by the stats. The Rockets really do have active hands at 5th in the league in deflections. So while the Rockets aren't contesting when the ball is shot, they're clearly making the defender uncomfortable in the process of getting the ball. Defensive rebounds. This was surprising to me. The Rockets are 3rd in the league in defensive rebounds by percentage. Two things here: One, it affects PPP with no easy putbacks. Two, it reduces the opponent's overall possessions and chances to score. The defensive rebound rate starts with Clint Capela, 4th in the league by defensive rebound % rate. But again, look down the list and you'll see Nene, Chris Paul, Harden, PJ Tucker, and Luc Mbah a Moute. Turnovers, Pace, and Transition points. From a holistic perspective, the Rockets high % ISO rate + slower pace on offense helps the other four guys get back on defense. Likewise, turnovers are down. This translates to fewer opponent points off of turnovers or in transition, shots with traditionally higher percentages. Five guys versatility. This year, there's no clear gaping hole for an opponent's offense to attack. As this 538 article states, "Of course many of these improvements, if not all of them, stem from the same thing: the club’s getting solid, versatile defenders in the off-season." It talks about how the interior bigs can come out, or the wings can guard multiple positions, "With smarter, more switchable defenders on the floor, the Rockets have also taken away many of the corner 3-point opportunities that were there for opponents in years past." Having five guys who can do multiple things on defense is paying off. In summary, Morey has had one of his most successful seasons ever, getting guys like Chris Paul, Luc Mbah a Moute, PJ Tucker, and Nene. These guys are smart, versatile defenders that integrate well into the Moreyball/D'Antoni system. Additionally, Clint Capela and James Harden have really stepped it up on the defensive end. Capela deserves a significant pay raise!
Carmelo Anthony's night recently is a prime example. They were weren't even in the neighborhood on any of his jumpshots. Absolute illusion of good defense. I don't think they coach detail that well. Luc is the only guy who keeps his hands high. Gordon typically makes athletic recovery contests but isn't disciplined overall. Everyone else is poor. There's no way interior defense should be so poor either. But they over switch and leave Harden/Paul/Gordon/Ariza at the rim as help defense. Capela can absolutely drop back in the pick and roll at a high level. I remember he and Luc having great defensive synergy playing this way. Does no one else remember all the steals Clint got in the pick and roll early in the season? I don't know why they have ramped up the switching so much.
So which role players are least expendable next season and if any are lost those spots need to be replaced with similar type players: lengthy, athletic, 3ball, high energy, D focused and hopefully younger than 30 (if younger than 25, may be cheaper) Are Rockets developing those type of players in G League now?
I think this might be intentional! The league average FT% is close to 70%. Maybe Moreyball and the coaching is telling the players to do anything on defense except cause a foul in the act of shooting. That would severely increase the opponents PPP, in the same way Harden does it on offense. Instead, disrupt the offense before the shot happens, so that if a foul occurs, it's not in the act of shooting. Is this giving Moreyball advanced analytics too much benefit of the doubt? I don't know but it's a plausible explanation.
Great observations. Two more things to watch: 1. Opponent FT rate--it's easy to be good when the other team doesn't get to the line. In the last 15 games, we lead the league in Opponent FT rate with a slender .202. (Of Course, our FT rate is .326 which doesn't hurt our scoring and control of the game at all.) 2. Rebounding--we started the year with this as a huge point of emphasis and it's carried over. Our defensive rebounding is night and day better than it was last year. And we get more than our fair share of offensive boards too. 3. Opponents FB points. Super low in the last 15 games (league leading 8.6).
I don't know, from what I've seen, it's because we switch so damn much that we eventually have to double which leads to open shooters
Good point, I was thinking steals and deflections reflect their run and gun style of offense for easy transition points. It's like their defense is still MDA inspired where good gambles leads to a payoff for the team and defender. Why not both?
In one part of this post you mention ppl talking about "some measure of time," but then you go looking at entire season measures. I really think this defense improvement is fairly recent, and your research might reveal something more if you filter by when Harden came back from injury....or at least when Luc came back from shoulder ... basically, when the team was together. I mean, why not? It's easy to do on NBA stats, if you're willing to accept less games as your research. We are indeed "amazeballs" in that time frame. Top 3 DRtg, right?
No, the defense has been inconsistent but started producing results before the recent win streak and Luc Mbah a Moute's return. The 538 article I linked to was written Dec.07, 2017 (when some of the stats I gave were then actually better) and this less detailed ESPN article is from January. If anything, this shows the defensive improvement came earlier in the season and that the Rockets have had slumps in their defense from early on.
I don't really care about so-called contributing stats as much as just pure DRtg. I agree, some of them are interesting, but I first look at just DRtg. We've been Top 5 DRtg for quite awhile, but not through December as much. Look at January 7th on. I'm interested mainly in explaining that, because that is what we are now, and it is the best 2 month period of our season, and it covers two different Paces, our regular Pace in January, and the slow Feb. I'm not so much interested in explaining Nov and Dec or the entirety of the whole ... since we are markedly better now.
As @heypartner said. I would just go by games which had Capela/Luc MaM/Ariza/Nene/CP3. Those players will hopefully be there for our playoff games. Our defense takes a hit when two of these guys are removed......and a top 3 defense when all are playing. More than just deflections, Def rebs, steals. Playing smart defense can mean leaving the poorest shooter open while sending 2 or 3 guys at a star......rotate appropriately. Thus the uncontested-late arriving 3 or uncontested late arriving mid range shot. The question might be who are the Rockets leaving open? A center who doesn’t shoot much? Andre Roberson types? This is why Pelicans role players nailed their 3-pointers. Those left open like Rondo and Moore made their shots.
