I am grinding my teeth to dust over this CSN Mid-Atlantic article about the “new heights” being reached by Scott Brooks’s offense in Washington. Author Chase Hughes is impressed because Washington’s offense has accomplished two things that were never accomplished by the Thunder during Brooks’s time in Oklahoma City: averaging 107.9 points per game; and scoring more than 100 points in 23 straight games. To Hughes, these numbers indicate that Washington’s players are thriving under Brooks to a degree that even the Kevin Durant-Russell Westbrook-James Harden core couldn’t in OKC.
What I am about to say, I say as a passionate, desperate Wizards fan: this analysis is silly and wrong.
Really this is a gripe about the kinds of very poor analysis that can still be published in the world of paid sportswriting in 2017. This particular analysis is so poor and poorly considered I feel stupid even rebutting it. Anyone who follows NBA basketball at all has the tools at their disposal to see where it obviously fails.
But, since we’re here, let us count the ways. No OKC team during Brooks’s tenure finished better than 3rd in points per game, but after 2009—OKC’s leap year, when they first became a playoff team—the team never once finished worse than 5th in points per game. This is notable because the apparently high-powered Wizards of 2016-17 are currently 10th in points per game. So, just this cursory glance at the most basic and useless stat in basketball tells us that scoring is up across the board in the NBA, such that a team in the top three in points per game in, say, 2013-14 wouldn’t even be in the top 10 in 2016-17.
Still, this stat alone doesn’t necessarily tell us much about the relative qualities of the teams being compared—it could still be possible that the Wizards are just better and more proficient offensively than those high-powered Thunder squads, in a league where maybe teams are just better at offense across the board. Maybe the whole league is better at offense!
Hmm. Let’s check on that. OKC’s 2013-14 squad—a 59-win team and the highest scoring OKC squad in terms of points per game of Brooks’s tenure there—scored about 106.2 points per game in the regular season, good for 5th in the league. But here’s an interesting stat: though that team scored a robust 108.1 points per 100 possessions, that rate—1.081 points per possession—was only good for 7th in the league that season. So how did they wind up with the 5th-ranked offense by points per game? Well, the Thunder played at a faster pace than most teams in the NBA, and so they had more possessions than many teams, and so their very good points-per-possession performance meant more total points in a game.
The 2016-17 Wizards also play at a faster-than-average pace, which means they, too, get more possessions than the average team. In fact, the Wizards play even faster than that 2013-14 OKC squad that won 59 games and had the 5th ranked offense by points per game! The 2013-14 Thunder averaged 97.91 possessions per game, good for the 9th-most possessions per game in the NBA that season. Not bad! But check this out: the 2016-17 Wizards average a whopping 99.41 possessions per game, good for an outrageous, mind-boggling, chart-topping, new-heights-reaching 12th in the NBA!
Wait, let’s look at that again: the 2013-14 Thunder had the 9th fastest offense in the NBA at just under 98 possessions per game, while the 2016-17 Wizards have the 12th fastest offense in the NBA at over 99 possessions per game. Huh. It’s almost like...the league...as a whole...is doing more...scoring?...because the game...today...is...faster?
Wait, that can’t be right. I thought we were talking about the “new heights” of Scott Brooks’s offense, not just that the entire league is using more possessions per game. Surely there must be a point of comparison that will strip away the noise of points per game and possessions per game and tell us which of these two teams was better at pure scoring, at converting offensive possessions into points. Just a simple number, maybe a points per possessions type thing.
Yes, hmm, it seems there is such a stat. It seems, in fact, that I have already referenced it in this very blog. What does it have to say about the scoring aptitude of these two teams? Well, the 2013-14 Oklahoma City Thunder scored 108.1 points per 100 possessions, and the 2016-17 Washington Wizards score 107.7 points per 100 possessions. So, actually, no, the Wizards are not more proficient, offensively, than at least this one Scott Brooks-coached Oklahoma City Thunder team. They are roughly half a point per hundred possessions less proficient, in fact.
For that matter, in terms of how dangerous the Wizards are on any given possession, they fare worse than all but two Thunder squads from the years when all of Scott Brooks, Russell Westbrook, Kevin Durant, and James Harden were on the team: the 2009-10 Thunder, when James Harden was a rookie and the Thunder were coming off a 23-win season; and the lockout-shortened 2011-12 season, when league-wide scoring was down so much that, by points per possession, the 10th ranked Los Angeles Lakers had about the same offensive rating (103.3) as the 2016-17 Los Angeles Lakers (103.2), who rank 25th in points per possession. In other words, the 2016-17 Wizards are less proficient on offense than any fairly comparable Thunder team meeting the criteria. Some of those Thunder teams, like the 2012-13 squad that won 60 games, absolutely tower over these Wizards (110.2 to 107.7) in offensive efficiency.
Look. None of these stats are perfect. Pace alone doesn’t describe how a team arrives at its level of possessions per 48 minutes—are you forcing stops and turnovers and taking advantage of transition opportunities, or are you taking bad early shots in half-court situations—and so it doesn’t describe whether an offense is good or reckless. Offensive rating is also incomplete, although, in basic terms, it does a good job of describing how dangerous a team is with the ball on a given possession. There is no single stat that says THIS TEAM RUNS BETTER OFFENSE THAN THIS OTHER TEAM. Even comparisons of two teams in the same season can be noisy and incomplete, which is why using a ridiculously flawed stat like points per game to compare teams from different years is absurd and pointless.
Here’s a quick way of making sense of that: in 2012-13, the fastest team in the NBA used 98.64 possessions per game, and the slowest team used 90.86. When the league’s pace is in that range, even a team inclined to play super fast is going to have a low ceiling on just how many possessions it can generate per game over the course of a season, even if it sells out on the offensive glass. Here’s an extreme example: let’s say one team plays at such a ridiculously fast pace that their possessions average only 8 seconds in length across an entire game. And let’s say their opponent plays at such a ridiculously slow pace that their possessions average a whopping 22 seconds in length across that same game. The fast paced team, despite only taking eight seconds per possession, still only gets 96 possessions in the game, a rate that would be 4th slowest in the entire NBA this season. This is basically what the Cleveland Cavaliers attempted in the 2015 NBA Finals, by the way.
Try to imagine what points per game, as a stat, tells us that is meaningful in a league in which the 19-42 Phoenix Suns, who have the 21st ranked offensive rating at 104.3 points per 100 possessions, have the 9th “best” offense by points per game, at 107.3. Points per game, without other numbers around it to twist it into shape, tells us absolutely nothing of value. The notion that it hints at the new heights of Washington’s offense is ridiculous. There are available numbers that say the Wizards are very good on offense, but there is no credible number that says they’re better than the Brooks-Durant-Westbrook-Harden Thunder. Arguing that point on points per game is announcing yourself as a goober, or a homer, or both.