Where Do James Harden and Dwight Howard Rank Among NBA's Top Tandems?

This summer James Harden made waves when he said that he and Dwight Howard were superstars, while the rest of guys were “role players.” Lost in the chagrin over his statement was the fact that Harden was right. Not only are they superstars, they’re also one of the best tandems in the NBA.

Where, do they fit, though, with the other stud parings?

In order to determine that, using the Play Index at Basketball-Reference.com, I looked at the top 100 players based on win shares, who had a player efficiency rating of at least 18 and averaged double figures in points. That narrowed the field down to 51 players.

(Note that because of Kevin Durant’s limited playing time, this excluded him and Russell Westbrook. That’s no reflection of my opinion of the pair, whom I regard as the best Batman and Robin in the NBA).

Then I eliminated players from losing teams or those who didn’t have a qualifying teammate. In some cases, such as the Cleveland Cavaliers and Dallas Mavericks, there were more than two players who met the criteria. In those instances I subjectively chose which player to use based on who had the better overall stats.

I ended up with 13 pairs. Here they are:

Team	Player 1	Player 2
Atlanta Hawks	Jeff Teague	Al Horford
Chicago Bulls	Jimmy Butler	Pau Gasol
Cleveland Cavaliers	LeBron James	Kyrie Irving
Dallas Mavericks	Dirk Nowitzki	Monta Ellis
Golden State Warriors	Stephen Curry	Klay Thompson
Houston Rockets	James Harden	Dwight Howard
Los Angeles Clippers	Chris Paul	Blake Griffin
Memphis Grizzlies	Marc Gasol	Mike Conley
New Orleans Pelicans	Anthony Davis	Jrue Holiday
Portland Trail Blazers	LaMarcus Aldridge	Damian Lillard
San Antonio Spurs	Tim Duncan	Kawhi Leonard
Toronto Raptors	Kyle Lowry	Jonas Valanciunas
Washington Wizards	John Wall	Marcin Gortat

The next thing I wanted to do is evaluate raw production. I added points, rebounds, assists, steals and blocks per game.

While as a stand-alone notion, box-score stats have their flaws, they certainly make for a good starting point. In this chart, the numbers are “stacked” so as to give a cumulative impression.

The reason for that is categorical comparisons by “winners” can be misleading. I don’t like arguments like, “Player X has 10 more points, but player Y has .10 more blocks, so each “wins” one category.” Stacking the columns accounts for aggregate differences. You can hover over each section, though, to see the actual numbers.

At least in terms of volume of production, the Houston Rockets’ duo is as productive as any on a per-game basis. Their total stats are 78.9, which is 7.2 higher than the second-best Trail Blazers. There are three teams—the Dallas Mavericks, Toronto Raptors and Atlanta Hawks—whose teammates didn’t reach 60 in total stats, so I trimmed them to form a list of “top-10 tandems.”

There still needed to be more refinement, though. The problem with this method is it treats every stat equally and doesn’t account for usage or shooting percentages. There are numerous single-number metrics which attempt to balance out such factors. Each has its own strengths and weaknesses.

Therefore, I looked at several of them. For all of these, I’ll keep the teams arranged in the same order as in the above chart for the sake of comparison. Click on the links to read more about them and the formula used to derive them.

The most common metric is Player Efficiency Rating (PER), which was authored by John Hollinger. It adjusts for minutes, and the formula is arranged such that it adjusts for what other players in the league are doing as well. Here’s how they stack up based on PER.

There, the New Orleans Pelicans tandem of Anthony Davis and Jrue Holiday stands out, but that’s mostly on the strength of Davis’ record pace of 33.0. Harden and Howard are fourth.

The next stat is win shares, which is compiled at Basketball-Reference.com. The problem with win shares is that it’s a cumulative stat. But not everyone has played the same number of games, so some players have advantages. However, adjusting it to per minutes distorts the advantages aggregated stats offer. Therefore, I used win share per game as a comparison.

The advantage to win shares is that the stat at least attempts to involve defense; though, some would argue it doesn’t do it well enough. Regardless, it’s another tool in the box—and a useful one:

The Rockets are fourth here, with the Golden State Warriors' Splash Brother’s, Klay Thompson and Stephen Curry first.

The next number is Floor Impact Counter (FIC) at RealGM.com, which gives greater weight to offensive rebounds and assists than the other metrics. Here’s a look at the twosome:

Based on FIC, Houston's "H" Brothers are fifth, and the Pelicans pals are first.

If none of those are the cake beneath your favorite brand of icing, how about a PIE instead? Player Impact Estimate is a metric from NBA.com/Stats which “measures a player's overall statistical contribution against the total statistics in games they play in.” It’s a bit different in that it’s varied for the pace of the game. It represents the percentage of events which transpire in a game which are recorded by the player in question.

Here are the partners in PIE:

The Rockets’ dominating duet leads the league in this metric.

Finally, if you want to see the overall snapshot, the following table indicates where each paring ranks in each of the aforementioned metrics. The final column denotes the average rank.

Pairing's Team	Total Stats	PER	Win Shares	FIC	PIE	AVG
Clippers	3	2	2	2	2	2.2
Rockets	1	4	4	5	1	3.0
Warriors	5	3	1	3	4	3.2
Pelicans	6	1	5	1	5	3.6
Trail Blazers	2	5	3	7	3	4.0
Cavaliers	4	6	7	6	10	6.6
Bulls	7	9	6	4	9	7.0
Grizzlies	8	7	8	10	7	8.0
Spurs	9	8	9	8	8	8.4
Wizards	10	10	10	9	9	9.6

In terms of averages, Los Angeles Clippers' cohorts, Chris Paul and Blake Griffin, are the most consistently near the top. While they didn’t lead in any one category, they were second in all the metrics and third in the total stats. Howard and Harden’s average rank is second-best.

Everyone is going to have their opinion about which stats they prefer to use. And, being perfectly honest, some “prefer” the stats which best promote their position. That said, regardless of which you use, the Rockets' superstars are in the conversation for best paring in the league (if you choose to go with FIC or total stats) and no worse than the fifth among those who have played significantly this year.

That’s enough to validate they’re the superstars. Does the fact that Harden was right get him a reprieve at all?