How Iso Melo was freezing Russ, George and the OKC offense

When Carmelo Anthony and Paul George joined Russell Westbrook in Oklahoma City this offseason, it looked like three superstars coming together to form a new superteam, an offensive juggernaut. But the Thunder are a dismal 17th in offensive rating.

One culprit: Anthony tends to play a lot of isolation basketball, preferring to have the ball in his hands to try to create scoring opportunities for himself off the dribble. This creates a lot of inefficient offense for OKC.

It's clear Thunder coach Billy Donovan recognizes the problem and has asked Melo to adjust to a new role in which he handles the ball less and works off the ball to get his shots. On Friday, Melo set season highs with seven made 3s on 12 attempts, mostly in catch-and-shoot situations.

"I think for me it's just a matter of accepting that role. That's all it is," Anthony said Friday night. "Realizing that's what it's going to be, these are the type of shots I'm going to get, this is the type of offense we're going to be running and accepting that, and working on that role. That's something that I've kind of been doing over the past week, is allowing myself to accept that role and do whatever I gotta do to make this team win."

The sooner that Iso Melo disappears for good and Anthony takes full advantage of playing with stars, the sooner the Thunder can reach their potential.

There are three key reasons why Anthony's iso game is a problem:

1. He doesn't create team offense

It has been demonstrated that volume-scoring wings like Anthony have a larger impact if they are also strong at creating team offense with the pass. Anthony's 17.4 assist percentage in his past four seasons with the Knicks was significantly behind that of Olympic teammates LeBron James (36.9 percent), Kevin Durant (24.6 percent) and Kobe Bryant (26.7 percent over his last four seasons). This season Anthony is down to a 7.8 assist percentage.

2. He creates primarily midrange jumpers for himself

The midrange jumper is the least efficient shot in basketball -- it goes in less often and draws fewer free throws than shots near the rim, but it's still worth only two points. This is the foundation for the Houston Rockets' philosophy of shooting either a 3-pointer or at the rim on almost every possession, and the same logic has permeated much of the NBA. Offenses built on iso play generate a lot of midrange looks off the dribble.

This is why Donovan has talked about reducing the number of "nonpaint 2s" the Thunder shoot, as reported by ESPN's Royce Young.

Anthony is one of the best iso scorers of his generation, but that doesn't solve the efficiency problem. According to Basketball-Reference, 47.1 percent of Anthony's field goal attempts this season are midrange 2s, barely down from his 49.5 with the Knicks last season.

By comparison, Durant took only 33.3 percent of his shots and James 23.4 percent of his shots in that range from 2014-15 thru 2016-17.

3. He doesn't imbalance the defense by involving multiple defenders

Some scorers get multiple defenders moving with post-up play that leads to double-teams or by running a lot of pick-and-roll plays. Others do so by driving to the basket. But Anthony engages in very few of these actions.

Anthony generates a lot of shots for himself, putting up points and maintaining "star" status. But at the team level, he's not generating star-level impact.

In Oklahoma City, that's especially damaging, because every possession Anthony uses is one that Westbrook, OKC's truly elite offensive engine, can't use to generate better results for the team.

Over the past three seasons, according to ESPN's real plus-minus (RPM) stat, Westbrook has been second in the NBA in each season in offensive impact. RPM estimates how a player impacts the team's scoring margin while he's on the court via a combination of plus-minus and box score stats.

This measure is broken down to offensive RPM (ORPM) and defensive RPM (DRPM). This is how the Thunder's three star players measured out in ORPM over the past two seasons, leading into this one:

Westbrook is on a different level from Anthony and George as an offensive force. Anthony, in particular, has not produced the level of positive impact that one might expect, even though offense is the primary way that he affects games.

Look at the shots Anthony is taking, according to Second Spectrum: 22.8 percent are pull-up jumpers, and 15.5 percent are "shake and raise," which means that a total of 38.3 percent of his shots are jumpers off the dribble. The pull-up jumper is the most frequent type of shot that Anthony takes, and he has just a 36.8 percent effective field goal percentage (eFG, which factors in the true value of 3s compared to 2s) on those shots.

Given that Anthony is playing with one of the greatest offensive engines in the game, there is no need for him to create so much for himself off the dribble. In fact, Anthony is an excellent catch-and-shoot player, either spotting up or catching the ball on the move. According to Second Spectrum, here are his numbers in those situations from both last season and this one:

So, even this season when his iso jumpers aren't falling, Anthony is still knocking down his catch-and-shoot opportunities at a high clip, like he does here:

OKC and Melo have recognized this, as Anthony acknowledged on Friday when he explained the change in shot selection: "Just talking, communication, watching film," he said. "With coach figuring out what this team really needs from myself, and I said something back in New York that we all need to figure out what our roles are on the team, so I think I figured out what my role would be on it."

In its last two games, OKC has posted an outstanding offensive rating in each: 119.2 vs. Utah on Wednesday and 124.2 vs. Atlanta on Friday. To keep the momentum going, OKC should continue to look for opportunities for Anthony to be a finisher, instead of a creator. As the Thunder showed on Friday, one key mechanism for doing that is to involve him in the on-ball pick action as the screener for Westbrook, usually one of the most active users of ball screens (along with former teammate James Harden).

In 2016-17, according to Second Spectrum, Westbrook utilized 65.7 percent of all direct picks used by the Thunder and generated 0.96 points per possession. In 2017-18 thus far, Westbrook has utilized only 53.3 percent of OKC's direct picks and generated 0.91 points per action.

But when Anthony is the pick-setter for Westbrook, that play has generated 1.031 points per direct pick (with a total of 130 direct picks). It creates easier shots for the players involved as well as for teammates in motion in off-ball action.

During the first two months of the season, the Thunder did not perform as expected. Some improvement should have been expected to happen organically, as they got used to playing with one another. More improvement should come when the Thunder's stars make more shots, as they have not lived up to their usual shooting percentages.

But just as important is that the Thunder's superteam experiment began as too much of an equal-opportunity offense when Westbrook is the only elite offense generator on the team. If Anthony can stay away from being Iso Melo and embrace the less glamorous role of screen setter and finisher, OKC's offense could improve dramatically as a unit moving forward.

Donovan sounded optimistic on Friday: "I give him a lot of credit because this is different for him, and he's really been open-minded about trying to do whatever he can do to help the team. And as a coach you really appreciate someone that's been in the game such a long time and established himself as a 10-time All-Star, 14 years in the league, one of the all-time greats, that he's willing to look at ways that he can do things a little bit differently to help our team."