2015-16 NBA Preview: Houston Rockets

HOUSTON ROCKETS

The league’s flagship analytics team, the Rockets have had success recently and they brought in another summer prize with Ty Lawson. They have a distinct playing style, eschewing midrange shots to an extreme degree while attacking the basket hard in transition and throwing up a huge volume of three’s. Houston has proven, along with the Warriors, that an extremely three-happy style can work at a high level. With a strong core and some depth, the Rockets look like a legitimate title contender.

2015 in review:

While the Rockets were beset by injuries as Dwight Howard was limited to half the season, Patrick Beverley missed 26 and all of the post-season, and Terrence Jones played only 33 games, the season was still a positive. The team was first in three-pointers attempted and made, naturally, but they were a better defensive club than offensive. They won 56 games, good for 2nd in the west, James Harden had a career year and came in a close second in the MVP race, and the team made it to the conference finals where they stole a game from the eventual NBA champions.

Rotation players in: Ty Lawson.

Rotation players out: Josh Smith, Joey Dorsey, Kostas Papanikolaou.

Houston made surprisingly few changes, but given their breadth of available talent and a deep playoff push no major overhauls were needed. Lawson was acquired cheaply in a trade, and if his personal problems are kept under control he’s a huge talent infusion. Josh Smith was a useful big man off the bench, often in unorthodox lineups, but with (hopefully) better health Houston won’t need quite so much depth. Both Kostas Papanikolaou and Joey Dorsey were parts for the Lawson trade; the latter was a backup center, but that’s ultra-athletic prospect Clint Capela’s job now.

2016 Projected

What’s important to understand about these projections is that most systems are not based on wins but something like point differential. The Rockets won 56 games, but they outscored teams by just a hair under 3.5 points per game, which is something you’d see from a 50 win team[5. The organization’s GM would agree with the method of focusing on point differential, not just pure wins, by the way.]. Thus, if a statistical prediction says 56 wins or something similar, it means the Rockets are actually improving.

Even without factoring in Ty Lawson, Houston should be better this season. The likelihood Howard will play more minutes is high, and he may even double his minutes. He could even play better this season since he showed renewed vigor on defense. But even a reduced Howard is a huge plus at the rim both from a rim protection standpoint and rim deterrence.

If Howard has trouble staying on the court, Clint Capela is ready to swoop in and block everything, catch lobs near the rim, and hopefully vanquish his free throw demons. He’s the rare player in that he’s tallied more minutes in the playoffs than the regular season, so be wary of certain metrics because his regular season stats were terrible. His playoff numbers probably aren’t a fluke though because his translated stats were so great from his time in other leagues — he had elite rates of blocked shots, rebounds, and field-goal percentage at a young age.

Rounding out the frontcourt, Ariza predictably couldn’t repeat his 3PT% in Houston but his defense was a vast improvement over Chandler Parsons’. He’s an ideal starter on a great team. But the most interesting piece, obviously, is Ty Lawson. Beverley as a dogged defender fits well next to Harden, but Lawson will need to start due to his pedigree. The thing to watch is how well the Rockets stagger their minutes so Lawson is always playing when Harden isn’t. That maximizes Lawson’s value because both guards are fairly high usage, high assist players[4. Shot volume and assists are on nonlinear scales. If you add a high volume shooter, the team doesn’t increase its number of shots; the proportions just change between players. It’s mostly the same with assists. Thus, if most of your value comes from handling the ball a lot, combined with a similar player you see generally see a decrease in your value.].

The most difficult adjustment is one welding Lawson and Harden together in the same lineup. Functionally, there’s only one ball, so either one or both players will see his stats drop, as was seen with LeBron James and Dwyane Wade in Miami in 2011. Overall, however, high usage lineups are generally successful, and since both players are good outside shooters they can at least space the floor for each other. Lawson and Harden also drive to the basket so frequently that it’s not really physically possible for them to sustain their previous rates in the same lineup, but they can at least draw offensive pressure and help one another, ideally.

With a strong starting lineup, Houston’s bench is still pretty strong with Beverley and Donatas Motiejunas providing solid production as well as Corey Brewer and his constant wind sprints adding fast break points. The team is still pretty young too, so there’s a chance one of the young guys has a breakout season. Finally, given the front office’s reputation, expect to see a clever move or two during the season and maybe one unorthodox player added who contributes to their bottom line despite a low prospect status.

Quick statistic/graph

Everyone knows that three-pointers are associated with statistical-savvy teams, but that’s not the only component those teams target. Since transition plays are generally more efficient, teams like Houston will push the pace and emphasize fast break scoring to increase these plays. Philadelphia famously does this with limited success because of their lack of talent, but for the past three seasons Houston has been doing this too. No team over that period has matched their tendencies of shooting from behind the arc and generating more of their points on fast break plays as defined by NBA.com.

Summary

Houston will be significantly better this season, and paradoxically that might mean they could equal their win total from last season. With some good health and a little bit of luck, they could have another deep playoff run too, and Ty Lawson provides the Rockets with another high-level shot creator and ball-handler. He can support the Harden-less lineups, which optimally would mean Harden could play less often and rest. With the other giants in the west, Houston needed a big move like this for the arms race. It most likely does not make them the best team, but it at least gives them an opportunity.

PBP-Metric[1. This is the initial version of my own metric, which uses a full range of stats collected from play-by-play logs and tested extensively to avoid overfitting.]: 55.2

PT-PM: 52

Nick‘s[2. For a short description, the predictions use regression models and neural networks to apply various stats like BPM, RAPM, and Win Shares to 10,000 simulations of the season game-by-game to select the “best” result.]: 57

Nathan Walker: 55