Nylon Calculus: Dissecting lost ball turnovers per 100 dribbles

CLEVELAND, OH - JUNE 07: Stephen Curry
CLEVELAND, OH - JUNE 07: Stephen Curry /
facebooktwitterreddit

A lot of the time, a seemingly “new” advanced stat is really just a slight modification of a more traditional formula. For example, take defensive rebounds per game, which has commonly been replaced with defensive rebound percentage. In these cases, the labels may be slightly altered, but the concept is not at all unfamiliar. It’s just a different denominator in the fraction — games played in one, rebound opportunities in the other.

When we add these more descriptive, more complete and more useful tools to our disposal, we open ourselves up to even greater possibilities. For example, using the tracking data from NBA.com/Stats along with the play-by-play data from Basketball-Reference, I was able to calculate lost ball turnovers per 100 dribbles. The results (for players who played more than 1000 minutes last season) can be found here.

After a quick glance at the macro level, the results seem reasonable. The point guard position is the most compact and contains the lowest values in the data set, while the center position is much more dispersed.

But before diving too deeply into individual players, it’s important to remember what this statistic is telling us — and what it’s not. The above table does not represent a ranking system of the flashiest ball-handlers in the league (just take a look where Kyrie Irving ranks among point guards). Lost ball turnovers per 100 dribbles is the rate that players (in their current role) turn the ball over while dribbling.

Here are a few more important points that need to be remembered when evaluating the above table:

  • Not every dribble is the same. For example, John Wall’s dribbles are likely more difficult than Otto Porter’s dribbles (whose ball-handling duties come mostly in situations when the defense is already at a disadvantage).
  • I have included positional estimates from Basketball-Reference to assist in sorting out the contrasting skills and tools of players at different positions. Although this still doesn’t account for specific roles within a position (like off-ball point guards compared to lead guards), the positional labels are intended to group similar players together to make analyzing the data easier.
  • These positional estimates (although useful for the above reason) are far from perfect. For this exercise, I placed a player in the position where they played most of their minutes this past season (so the positional label for a few players may be surprising).
  • A player may look proficient at a “higher” position, but much less capable at a “lower” position. For example, Patrick Beverley ranks in the 97th percentile as a shooting guard. However, move him to point guard and his rate looks a lot more unfavorable (43rd percentile).

Observations

Stephen Curry and Kyrie Irving, who both boast some of the NBA’s most captivating on-ball wizardry, ranked below average compared to their counterparts. Meanwhile, young projects like Emmanuel Mudiay and Dante Exum unsurprisingly posted two of the worst marks at their position.

Knowing that the primary playmaking responsibilities are generally delegated to their teammates, it makes sense that “ball-moving” point guards like Matthew Dellavedova, George Hill and Patty Mills dominated their counterparts at the point guard position.

Just as one would expect, catch-and-shoot specialists J.J. Redick, Kyle Korver, Allen Crabbe and Troy Daniels did not fare well.

Young perimeter players like Jaylen Brown, Andrew Wiggins, Devin Booker, Buddy Hield, Kentavious Caldwell-Pope, Kelly Oubre Jr. and Victor Oladipo still have work to do, all ranking in the 32nd percentile or below at their position.

Kevin Durant (39th percentile among small forwards) and Paul George (47th percentile) oddly struggled, while other wing greats like LeBron James, Gordon Hayward, Kawhi Leonard, Jimmy Butler and Giannis Antetokounmpo all exceeded the 70th percentile at the position.

Speaking of Kevin Durant (whose handle looks a lot better at the power forward position in this exercise), he crushed his mark from the season prior and posted his best number in the past four years. This breakout coincided with more minutes at the power forward position than ever before — an adjustment that almost certainly increased his opportunities to make plays.

In terms of the Warriors as a whole, Andre Iguodala, Patrick McCaw and Draymond Green all performed well. These are certainly talented offensive players, but Golden State’s exceptional spacing and assortment of playmaking options gives them a boost as well.

The ability to dribble handoff is an overlooked component in the modern offense, and experts at this skill (players like Nikola Jokic, Kelly, Olynyk, Mason Plumlee, Cody Zeller and Al Horford) all ranked highly.

Players who play up a position regularly like Harrison Barnes, Rondae Hollis-Jefferson, James Johnson, Kevin Durant, Tobias Harris and Danilo Gallinari receive a significant bump at the higher position, while Al-Farouq Aminu (95 percent of minutes at power forward this past season) still disappoints (37th percentile).

Other “tweeners” who rarely played up a position last season (Juan Hernangomez, Serge Ibaka, Jerami Grant, Rudy Gay, DeMarre Carroll, Luc Mbah a Moute, Carmelo Anthony and P.J. Tucker) performed poorly. Meanwhile, despite going through a seemingly foolish experiment where he played most of his minutes at the small forward spot, Aaron Gordon found himself in the 63rd percentile at the position.

Robert Covington, Timothe Luwawu-Cabarrot, Gerald Henderson and Nik Stauskas all struggled mightily as their playmaking weaknesses were amplified by the limited creation around them in Philadelphia. The Markelle Fultz and Ben Simmons duo can’t get here soon enough.

Next: Joel Embiid vs. greatest rookie center seasons of last quarter century

Lost ball turnovers aren’t the only type of turnover documented in Basketball-Reference’s play-by-play data. Another study like this one with bad pass turnovers could be in order as we continue to look for ways that film study and data analysis can come together to facilitate a deeper understanding of a player’s strengths and weaknesses.

Sources: NBA.com/Stats, Basketball-Reference