Nylon Calculus: Is the NBA really a point guard dominated league?

We often hear the NBA is a point guard dominated league and looking around at all the talent at that position — Stephen Curry, Russell Westbrook, Isaiah Thomas, Kyle Lowry, it’s hard to argue. But do the numbers actually bear his out, do point guards really run the world? We can look at the median, range and variance of a number of all-in-one metrics for each position to determine which position is currently the strongest in the NBA.

We’ll look at the following all-in-one metrics: PER, BPM, WS/48, RPM, DRE, HBox, Dredge, and PT-PM (minimum of 500 minutes played, PT-PM is updated through 1/7 while Dredge and HBox are updated through 1/29).

As we can see here, center seems to routinely show up as the strongest position no matter which metric we look at (the median — the line in the box — is always the highest). However, while the median is the highest for the center position, we can see that the range always seems to be the largest for the point guard position (the box and whiskers portion). This would indicate that there’s a big difference in point guard play throughout the league. And while the median point guard may not be as good as the median center, the top end point guards are generally better than the top end centers. So there is a case to be made that the NBA actually is a point guard dominated league even if on average, the centers are better.

Which position is currently the worst? When looking at the graphs, it appears the shooting guard position is the weakest with smaller ranges or a smaller median than most of the other positions (with the exception of Player Tracking Plus-Minus).

We can also notice some subtle difference between the metrics by looking at these graphs. For example, one of the reasons the Point Guard position might look stronger according to HBox is because the metric has many different interaction terms that use assists. For RPM, height is a variable and that’s why you might see the bigs with a higher median (when I ran this study a few years ago, I noticed the same thing — bigs tended to have a higher median RPM).

Finally, we can take note of the outliers (the dots on each graph). We can see that for centers, there aren’t any outliers with the exception of DeMarcus Cousins in HBox (Boogie has the highest AST% for Centers so this isn’t surprising).

For point guards, the outliers are generally one of Chris Paul, James Harden or Russell Westbrook. While Westbrook and Harden are putting up eye-popping numbers and are the two favorites for MVP voting, Paul has quietly had an excellent season himself — among PGs, he ranks fourth in PER, second in BPM, first in RPM, first in DRE, third in HBox, second in Dredge, first in WS/48 and first in PT-PM . Neither Westbrook nor Harden lead in as many metrics as Paul does (Westbrook is in first twice, Harden once and Lowry once). Perhaps, Chris Paul ought to be getting more consideration for MVP, injury aside.

For shooting guards, Ben McLemore is an outlier in the sense that he has performed so terribly On the positive side, Giannis is the one player who consistently shows up as an outlier among shooting guards (I used Basketball-Reference positions and Giannis is listed as a shooting guard where he has played 63% of his minutes. I don’t really buy that, especially because Tony Snell has started the majority of Bucks games and is listed as their SF according to Basketball-Reference. But I wasn’t about to subjectively change positions so I left Giannis as a shooting guard.) Patrick Beverly shows up as an outlier in Dredge (yes, he is having an underrated season). For small forwards, the outliers are the usual suspects: Lebron, Kawhi, KD and Jimmy Butler.

There are a few more bad players who show up as outliers when looking at BPM for power forwards: Jarell Martin and Noah Vonleh. On the plus side, Draymond consistently shows up as an outlier while Kevin Love and Paul Millsap make an appearance as an outlier in HBox and RPM respectively.