Nylon Notebook: Ball-movers, black holes and the Warriors blocks

DENVER, CO - DECEMBER 20: Andrew Wiggins
DENVER, CO - DECEMBER 20: Andrew Wiggins /
facebooktwitterreddit

We’re back with another installment of Nylon Notebook, looking at a few different statistical storylines from around the NBA. All statistics are current through Dec. 20.

Driving to shoot and driving to pass

Dribble penetration, either in isolation, from dribble hand-offs or pick-and-rolls, is an essential element of any offense. It gets the ball close to the basket, creates high-value trips to free throw line, and collapses the defense forcing rotations which can lead to open shots. The players who actually get the ball into the teeth of the defense often have strong tendencies about what they do with it once they get there.

I thought it would be interesting to break that down a bit, and look at players who drive to create shot opportunities for themselves versus players who drive to create opportunities for others. Among players with at least 150 total drives on the season, the players below have the highest percentage of their drives end in true shot attempts (either a field goal attempt or a shooting foul).

There aren’t a lot of surprises here on this list. Jackson ranking fourth is a bit interesting since he was billed as potential playmaker on the wing in this year’s draft. He’s recorded just four assists on his 160 drives this year. That balance wouldn’t be quite so troubling if he wasn’t also shooting 34.5 percent on drives, the second-lowest field goal percentage among the 92 players with at least 150 total drives.

And here’s the other end of the list, the players who are least likely to create a scoring opportunity for themselves off a drive.

Two of the top three happen to play for the Utah Jazz, which speaks to the design of their offensive system. Ingles is just a secondary playmaker but both he and Rubio are driving to initiate a series of ball reversals and accompanying defensive rotations that drain the shot clock (and hopefully defensive energy) before creating an opening on the perimeter. It’s also interesting to see both Lowry and VanVleet on the list, perhaps a testament to how much the Raptors have focused on adjusting their offense towards getting the ball back out to open shooters.

The Black Hole Factor

We’re all familiar with the concept of a ball-stopper or a black hole — a player who rarely passes it back to a teammate after receiving it. Building off the first section and the idea of players who drive to look for their own offense, I thought it might be fun to try and identify some true ball-stoppers.

Usage Rate and field goal attempts (in some ratio to assists) are often the most basic measures used to define ball-hogginess but we have the problem of interior players like Clint Capela or DeAndre Jordan, or catch-and-shoot specialists like Kyle Korver, who don’t touch the ball a lot but are supposed to shoot as soon as they touch the ball.

To try and account for them, I looked at the 279 players who have touched the ball at least 500 times this season, per NBA.com. I charted each player by their average dribbles per touch and the percent of their touches that end a possession. The latter figure is calculated as the percent of a player’s touches that don’t result in a pass. That means it does catch some unintended outcomes like non-shooting fouls drawn or holding the ball at the end of a quarter, but I think it suffices for this simple analysis. The dribbles measure is meant to highlight players who finish possessions and do it by pounding the ball for a relatively long period of time, as opposed to the Korver’s and Capela’s of the league.

At the top left you see low-usage point guards like T.J. McConnell and Jose Calderon who dribble a lot but rarely finish possessions. Players like John Wall, Russell Westbrook and James Harden all dribble the ball a ton but they also pass the ball plenty. For reference, Harden’s mark of 38 percent of his touches ending a possession is only the 89th percentile for this group.

The median number of dribbles per touch in this same of 279 players was 1.37. So the players who really could be considered “black holes” would be those on the outer edge from DeMar DeRozan to around T.J. Warren. While “black hole” and “ball-stopper” have negative connotations, that’s not necessarily the case. Of the seven players in that group who are labeled on the graph — Booker, Oladipo, DeRozan, Gordon, Fournier, Wiggins and Warren — only one (Warren, -0.64) has a negative Offensive Real Plus-Minus this season.

Not to pile on Andrew Wiggins but this does seem to be another problematic angle on his production. His ORPM is the second-lowest of that group — at +0.26, it’s essentially the same as an average player. Wiggins was theoretically supposed to become a lower-usage player this season, with Karl-Anthony Towns’ ascendence and the arrival of Jimmy Butler making him a efficient complementary piece. Wiggins hasn’t deferred that much. He’s touching the ball slightly less often per minute, but the nature of those touches hasn’t changed a ton — the length of his touches and his average dribbles per touch are pretty similar to last year.

Ed. Note — Our own Todd Whitehead also got a sneak peak of this graph and was interested enough to create his own variation. That graph is below along with Todd’s explanation.

Hopefully the chart is self-explanatory and correctly defines the meaning of a black hole – a guy who does more for himself than he does for his teammates. The formula I came up with “bakes in” the volume of touches by using the difference between “yours” and “mine” rather than the ratio. The only tricky part is how to treat the touches that are NOT: FGAs, FTA Trips, TOV or Potential Assists, i.e., the non-assist swing passes.

I used NBA.com team tracking-passing data to calculate the average number of passes per possession (3.02) and the average number of potential assists per possession (0.45) to find the average number of non-assist swing passes per possession (2.58). Then I created something I’m calling an “assist-equivalent swing pass” which equals the number of swing passes divided by 2.58. The idea is that each swing pass has some value (less than an assist but more than 0) as it will, in theory, lead to a better scoring opportunity later in the possession. I think it’s a reasonable approach. 

Next: Nylon Calculus -- More on the NBA Draft and why the tallest men can't dribble

Warriors getting blocks

A lot has been made about the Warriors’ ridiculous block numbers. Through Wednesday night’s games they were averaging 8.5 blocks per game, the third-highest mark since blocks began being recorded during the 1973-74 season. Obviously, this is remarkable because the Warriors often play without a traditional center. What’s even more incredible is that they’ve put up these numbers while avoiding fouls at record rates.

The Warriors are averaging 2.4 personal fouls per blocked shot, the lowest mark of any team since the 1973-74 season. The graph below compares them to the other 203 teams that have averaged at least 6.0 blocks per game over the past 44 seasons.

Only seven other teams have finished the season with fewer than 2.75 personal fouls per blocked shot. The Warriors are excellent at avoiding fouls in general but particularly around the basket, and especially Kevin Durant and Draymond Green. Together, those two average 5.1 blocks per 100 possessions to 7.6 personal fouls, a ratio of about 1.5 fouls per block.

Everywhere you look, it seems like the Warriors are threatening to make history.