The Turnoverinator

The NBA’s best at eliminating turnovers

In no uncertain terms, a turnover is just about the worst thing a player can achieve on a basketball court. It always leads to zero points, the opposition gets the ball, and teams are generally more efficient following a forced turnover. In terms of pure point swing, turnovers are deadly. The league in non-turnover offensive possessions has an offensive rating of 136.1 (according to my calculations), and teams have averaged 1.24 points following a turnover. Functionally, an offensive possession worth 1.36 points turns into a possession worth 1.24 points for the opposition. That adds up to a swing of 2.6 points, which sounds as bad as it is.

*All stats are as of November 18th

With this in mind, players who limit turnovers are incredibly valuable. It’s the rare offensive skill that substantially aids both offensive and defensive efficiency, and in any close game, the turnover battle is usually the deciding factor. However, basketball isn’t about limiting turnovers. It’s about getting buckets (or stopping them), and the players who are the best bucket getters, through scoring or passing, are also generally the highest turnover players. In many respects, turnovers are the cost of generating offense, which makes the players who shoulder a massive offensive burden, while simultaneously limiting turnovers, the best offensive players in the league. But who exactly are these offensive dynamos?

Fortunately, there is an incredibly strong correlation between individual turnovers and raw offensive usage. Raw offensive usage is the sum of a player’s true shot attempts and assists (AST). True shot attempts (TrSA) are calculated as follows: FGA + Shooting Fouls Drawn - And-Ones. This season, the correlation between raw TrSA + AST and TOVs is a staggering 0.8998. I cannot stress this enough: when you use basic basketball stats, it’s extremely difficult to find a correlation this strong.

Could I make this model slightly better? Yes, I think factoring in 3-point attempts would make it more accurate, but I’m trying to keep this as simple as possible. Due to the strong correlation between raw usage and turnovers, I was able to use a linear forecast model to spit out a player’s expected turnovers based on raw usage. From there, I subtracted their actual turnover figures to find the best players at truly limiting turnovers.

*Because limiting turnovers is a positive outcome, I wanted the figure to be positive as well

Honestly, Shai Gilgeous-Alexander’s turnover suppression might be his greatest superpower. He leads the league in raw offensive usage and is somehow only 74th in total turnovers. That is insanity. He’s an outlier like none other, and he isn’t even a 3-point specialist like many of the top-25. Another player who deserves a shoutout is Derrick White. Despite suffering through a horrible shooting slump, White’s ability to limit turnovers has been masterful, and it’s why he can have an Offensive Box Plus/Minus of 1.4 despite a true shooting percentage of 47.9%. When White finally starts hitting shots, he could be in for a monster year, particularly among the nerd-metrics. This list includes a bunch of real on-ball stars and a bunch of guys who catch-and-shoot threes. I’d be willing to bet that if you removed catch-and-shoot threes from True Shot Attempts, you’d get a more robust linear correlation.

Now that we know who the best players are at limiting turnovers, let’s take a look at the other end of the spectrum.

This is an interesting cohort of players. First and foremost, the centers who rank highly (lowly) are almost certainly being penalized for illegal screens. Next, people need to pump the brakes on Stephon Castle. He’s an interesting young guard, but he might be the most turnover-prone player in the league. That isn’t unsurprising or even problematic long-term, but right now, he’s playing well below his assist and point figures would suggest. And finally, this is another reminder that Devin Booker really benefits from playing alongside a solid point guard. He’s coughing the ball up more than you’d like, and it’s why he’s averaging 3.8 turnovers per game, his most since 2019-20, the season before Chris Paul showed up in Phoenix.

Hopefully, over the course of the season, I’ll throw in a few more variables to build a more accurate model. However, for the time being, just know that turnovers are bad, and the guys who limit them compared to their field goal attempts, assists, and free throw attempts are good.

For any inquiries about work, discussion, and the like, you can email me at nevin.l.brown@gmail.com.