Nylon Calculus: A look at player consistency

LOS ANGELES, CA - FEBRUARY 17: Devin Booker /

We often hear the phrase: “so and so is one of the most consistent players in the NBA” and yet we rarely see it quantified. A few years ago, we at Nylon Calculus decided to change that by introducing player consistency (the methodology and explanation of the metric can be found here and I suggest reading to get a better understanding of the metric and what it’s measuring). With the All-Star break finishing up, I thought it would be a good time to take a snapshot of the most consistent players in the NBA.

Player Consistency can be calculated for any number of stats from basic points per game to something more complicated like DRE (Daily RAPM Estimate). Since we want to measure a player’s overall consistency, I looked at four different linear weights metrics that attempt to capture the total value of a player within a single game: DRE, John Hollinger’s Game Score, Nathan’s SPR and David Lewin and Dan Rosenbaum’s Alternate Win Score (AWS). I also computed the average consistency of the four metrics which was then converted into a percentile (higher means more consistent). You can see the full results in an interactive table here.

Thunder rookie Terrance Ferguson has been the most consistent player in the NBA this year (sorting by Average Consistency). That’s right, a 19-year-old rookie has been the most consistent player. You’ll also notice rookies Malik Monk and T.J. Leaf among the most consistent players. This is surprising: conventional wisdom says that the more experience you have, the more consistent you’ll be. It’s not as shocking in the case of these rookies as they’ve consistently struggled. In addition, I checked the correlation between consistency and age and it was non-existent.

Among the star players, Boogie and Anthony Davis were the two most consistent stars followed by Joel Embiid. This just further illustrates how devastating the Boogie injury was: he was consistently great and the Pelicans could rely on a base level of performance from both of their stars in every game. As for Embiid, his consistent greatness is one of the reasons why he’s in the 99th percentile in on-off net differential.

On the other end of the spectrum is Klay Thompson, who is the most inconsistent player in the NBA. Considering Klay is playing with two MVPs, it’s not surprising to see he’s wildly inconsistent, as his opportunities to contribute vary dramatically from game to game.

Other inconsistent players include Eric Bledsoe, who’s change in teams might have something to do with that (we’d need to separate out his consistency scores with the Bucks and Suns to see if there’s been a change), John Wall, Jamal Murray, Andrew Wiggins, J.J. Redick and Goran Dragic. Wall’s inconsistency might help explain why the Wizards never really put together a string of great games when he was healthy while Redick’s inconsistency is an interesting contrast with Embiid’s consistency. Does Redick’s inconsistency have to do with the fact that he’s primarily a 3-point shooter? There are quite a few 3-point shooters who show up as among the least consistent players. Specifically, the lower usage 3-point shooters like Danny Green or Josh Richardson. It’d be interesting to look into whether the lower usage 3-point shooters are less consistent in overall impact but that is a study for another time.

Finally, I thought I’d finish this post by comparing players’ consistency with their overall value (I’ll use RPM to measure this). Again, remember that consistency isn’t necessarily good or bad but it can have an impact on the teams’ performance, specifically with regards to the teams’ best players (another potential study: what’s the relationship between team consistency and the consistency of the best player). You can see the full interactive chart here.

As can be seen here, there’s no relationship between RPM and consistency. There are really good players who are really inconsistent while there are bad players who are really consistent. Of course, for the top teams, it might be ideal if your players are both consistent and great but as we can see on the graph, both the Rockets and Warriors have star players who range from very inconsistent (Klay) to very consistent (Capela, Steph). Interestingly, despite having one of his worst seasons in his career from the 3-point line, Eric Gordon has been a fairly consistent player relative to the rest of the NBA.

Next: Nylon Calculus Week 18 in Review -- All-Star Weekend 2018

On the other end of the spectrum, the Suns best player Devin Booker, has been very inconsistent. A quick check of Booker’s splits reveals that he has been excellent in the Suns wins (59.6 true shooting percentage, while averaging almost 28 points per game) and not quite as good in their losses. This might help explain why the Suns have over-performed their expected win total: their SRS is by far the worst in the league and yet they remain cluttered with all of the other tankathon teams in a race to the bottom.

Booker’s inconsistency, which can lead to more peaks and valleys might actually be helping the Suns win some games because he’s so good in the teams’ wins. Of course, it’s possible this is all noise and there really is no relationship between Booker’s inconsistency and the Suns over-performance. This would be something to explore in the future: do teams who have inconsistent stars over and under perform their records more than those with consistent stars?