Did the James Harden And Chris Paul Experiment Work in Houston?

When the Rockets unexpectedly acquired Chris Paul via trade in the summer of 2017, the NBA world was abuzz.  How could the ball-dominant Paul coexist with usage king James Harden?

I was actually already thinking of writing about this, but now, with the Paul era in Houston coming to a premature end (hello Russell Westbrook!), I think its worthwhile to look back and evaluate the Harden-Paul pairing .

The Landscape Before Paul Went to Houston

First, a bit of context.  The 2016-2017 Houston Rockets, the edition before Paul’s arrival, were a very strong team.  They won 55 games and sported a +5.5 net rating, good for 3rd in the NBA, per basketball-reference.  They had a really good offense (James Harden was an MVP candidate, after-all) but a mediocre defense that was 0.2 points per 100 possessions below league average.

Harden, in his age 27 season, was already the free throw machine that we know and love with over 10 FTA per 36 minutes, per basketball-reference.  But this was the season that he upped his three-point attempt rate to 49%, from roughly 41% in 2015-2016.  The Rockets outscored their opponents by +6.4 points per 100 possessions with Harden on the court but were a more pedestrian +2.4 with him off, per pbpstats.  The decline was mainly due to the offense coming back down to earth without Harden on the court.

In Los Angeles, Chris Paul had a very fine statistical 2016-2017 season.  At age 31, he posted a career high 61.1% true shooting percentage, buoyed by shooting a career high 41.1% on three-pointers with a 38.5% three-point attempt rate (per basketball-reference).  His Clippers posted a +4.5 net rating that season, good for 4th in the NBA.  Continuing a trend for the lob-city era Clippers, the team was roughly 19 (!) points per 100 possessions better when Paul played than when he sat, per pbpstats.  Still, perhaps foreshadowing what was to come, CP3 played in only 61 games in 2016-2017, down from suiting up for at least 74 games in each of the previous two campaigns.

Did The Paul Experiment Work?

So, did it work?  Could Houston fully realize the benefits of having two superstars who thrive with the ball in their hands?

In short, yes, it essentially worked.  But with some caveats.

It is worth remembering that the 2017-2018 Rockets posted the best net rating in the NBA, at +8.6!  They were bolstered by the new and improved 6th ranked defense in the association.  They went from being 0.2 points per 100 possessions worse than league average in 2016-2017 to 2.5 points per 100 possession better than league average, per basketball-reference.  The 2018-2019 squad was not as dominant in the regular season as the 2017-2018 one was, but still managed the 5th highest net rating in the NBA at +4.8.

To see just how Harden and Paul made it work, it is instructive to look at their on-off stats for the combined 2017-2018 and 2018-2019 seasons.  I should note that here I am filtering to only the games in which both players played, to better highlight their effect together.  My source is the excellent basketball statistics and research website, pbpstats.com.

Screen Shot 2019-07-13 at 10.15.32 AM

While Harden was on the court, with and without Paul, the Houston offense thrived. But what is fascinating to me about these numbers is that 3rd row, the 1168 minutes where Chris Paul played and James Harden sat.  Keep in mind that these are for games in which both players played, so here Paul is leading bench heavy units at the end of the 1st and 3rd quarters and beginning of the 2nd and 4th, while Harden takes a breather.

In those Paul only minutes, the Rockets’ defense was really good, posting a 103.2 defensive rating! And the 115.7 offensive rating shows that Houston was able to keep its offense afloat with Paul and without Harden.  Both numbers are very impressive. For context, NBA offense, in general, tends to be less efficient for both teams when bench heavy lineups take the floor.  This makes the Paul only offensive rating even more important. Not only did Houston’s Paul-led units lock down their opponents, but they were also able to keep putting the ball in the basket at an efficient clip.

The Rockets were able to accrue these minutes with the bench unit plus Paul by effectively staggering their stars.  They typically took Paul out a little earlier in the 1st and 3rd quarters and then put him back in at the end of the quarter when Harden was taking a breather.   You can nicely see this pattern in this excellent visualization of NBA substitution patterns by Alex Wainger.  Utilizing this stagger, the Rockets played only 171 minutes over the two seasons without both Harden and Paul, for those games in which both played. 

Of course, Houston’s on-off stats with various combinations of Harden and Paul cannot be looked at in a vacuum.  In basketball it can be hard to disentangle individual credit from team success; the supporting casts contributed mightily to these numbers.  The 2017-2018 squad relied on bench contributions from P.J. Tucker, Luc Mbah A Moute, and Eric Gordon, while the 2018-2019 bench minutes featured more Gerald Green, Austin Rivers and Gordon.  The 2017-2018 bench players in particular deserve a lot of praise for Houston’s stellar season.

In terms of individual statistics, James Harden raised his high usage, high efficiency high-wire act to new heights in the past two seasons.  An over 61% True Shooting Percentage paired with plus 40% Usage speaks for itself!  Chris Paul, on the other hand, had an efficient scoring season in 2017-2018 but experienced a drop-off in 2018-2019.  We can see this in his declining true shooting percentage.  His usage also dipped a little, particularly in 2018-2019.  Data in the graphs below is from basketball-reference.

Screen Shot 2019-07-13 at 11.34.47 AM

Screen Shot 2019-07-13 at 10.58.46 AM

Paul still posted an above average true shooting percentage even in 2018-2019; league average is roughly 55%.  It should also be noted that at least some of the decline in scoring efficiency for CP3 was due to poorer outside shooting; his three-point percentage declined from 38.0% in 2017-2018 to 35.8% in 2018-2019.  And he hoisted over 6 three-point attempts per game in each of his seasons in Houston, the two highest marks of his career.

No evaluation of the Harden and Paul experiment would be complete without mentioning the fact that CP3 played in only 58 games in each season due to injuries.  A potential source of worry, Paul’s health, did in fact hamper the team. This was particularly true in the 2018 playoffs.  We all still remember Houston going up 3-2 in their series with the Warriors, before Paul had to sit out the remaining two games with a hamstring injury and Golden State ultimately advanced to the Finals.

Though I have to to do more research on this in general, I think the Paul experiment in Houston is a piece of positive evidence in favor of staggering star players, especially ball-dominant ones.  A player who can effectively create efficient offense on the second unit can be extremely valuable for bolstering the offense at a time when it is typically not as strong.  This piece of advice could be relevant to the Rockets again this year with the high-usage Russell Westbrook now paired with James Harden.

The Rockets did not win a championship in either of their seasons with Chris Paul, so perhaps history will judge this experiment as a flop.  Still, I think there was a lot to like from two seasons of one of most interesting superstar combos that the league has seen.



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: