The Worst On/Off-Court Numbers from 2014/15

Apr 29, 2015; Memphis, TN, USA; Memphis Grizzlies forward Jeff Green (32) celebrates with guard Courtney Lee (5) after a play against the Portland Trailblazers in game five of the first round of the NBA Playoffs at FedExForum. Memphis defeated Portland 99-93. Mandatory Credit: Nelson Chenault-USA TODAY Sports
Apr 29, 2015; Memphis, TN, USA; Memphis Grizzlies forward Jeff Green (32) celebrates with guard Courtney Lee (5) after a play against the Portland Trailblazers in game five of the first round of the NBA Playoffs at FedExForum. Memphis defeated Portland 99-93. Mandatory Credit: Nelson Chenault-USA TODAY Sports /
facebooktwitterreddit
Apr 29, 2015; Memphis, TN, USA; Memphis Grizzlies forward Jeff Green (32) celebrates with guard Courtney Lee (5) after a play against the Portland Trailblazers in game five of the first round of the NBA Playoffs at FedExForum. Memphis defeated Portland 99-93. Mandatory Credit: Nelson Chenault-USA TODAY Sports
Apr 29, 2015; Memphis, TN, USA; Memphis Grizzlies forward Jeff Green (32) celebrates with guard Courtney Lee (5) after a play against the Portland Trailblazers in game five of the first round of the NBA Playoffs at FedExForum. Memphis defeated Portland 99-93. Mandatory Credit: Nelson Chenault-USA TODAY Sports /

On/off-court numbers, available on the NBA.com stats page for every team, to me are some of the best numbers for analysis available. Typically, if there’s something interesting to be found about a player’s development or contributions on the court, on/off numbers are the first place you’re likely to find something interesting– if ever there is something interesting to be found. It’s the first sniff test for any piece of basketball analysis, in particular for conclusions that haven’t yet reached the general NBA audience (or even the nerdiest communities). Adjusted plus-minus systems go a long way to cancelling out noise and separating whether a player should be credited for a team’s good/bad performance while he’s on the court, but due to variance and multicollinearity issues trustworthy numbers may take a while to produce.

Previously, I’ve written about surprising plus-minus stars, and the difference between individual rebounding rates and contributing in a positive way to a team’s rebounding totals. These exercises have produced relatively interesting results. This time, I went looking for the opposite of surprising on/off-court stars. Players who (to an extent) may look good by traditional box score metrics whose numbers didn’t translate to team success last season. The hypothesis being if you’re helping your team it will show up somehow with the team’s performance with or without you on the floor.

This article would have been called the ‘Jeff Green All-Stars of 2014-15’, if not for the fact that I had used that phrase to describe J.J. Hickson’s impact on team rebounding already. But any evaluation of this type has to begin with pointing out Jeff Green’s career on/off-court numbers. Green’s teams have always, without a single exception, always played better with him off the court, and most of the time by significant margins. A fact that can’t be ignored and should tell you that Green isn’t the type of player to help a good team, at least given any role resembling the ones he has played so far in his career.

For this list, I’ve chosen players that stand out on their particular teams. In a way that makes you go ‘oh that’s not good at all” at first glance of after thinking about it some more. Note that if you do this exercise, there may be a case or two where you’d add/remove a player.

On/Off-Court Duds from 2014-15

*minimum 700 minutes played. Click on column to order by.

Three of the more interesting players on the list are Jeff Green, Tobias Harris and Channing Frye– each made their respective offenses better, but significantly worse on defense. In the case of Harris on Frye their absolute differential isn’t that large, but the two did post the worst numbers on the Magic. And at least in Harris’ case you’d expect him to be playing with the best lineups.

By far the surprise of the group is Jonas Valanciunas (and I can’t stop checking this). Valanciunas was the only Raptor (who played any real minutes) to end up with a negative scoring differential. Valanciunas ranked 30th in RPM at his position out of 62 qualified players. The Raptors were 8.3 points per 100 possessions better with him off the court, equivalent to the difference between the Clippers and Pistons in net rating last season. This fact amazes me and it escapes how it’s possible. Something certainly to keep an eye on this season.

It can be argued in the cases of Austin Rivers, Chris Kaman or Leandro Barbosa that their replacements (or rather players they were replacing coming off the bench as subs) were so good that their numbers are bound to be bad. This is true. However, each of these players still stood out significantly compared to any of their teammates. For Jarrett Jack, these numbers should shatter any illusion about him being a starting level point guard. Kevin Seraphin has been a favorite “underrated” mention for some analysts, but to this point in his career neither adjusted nor pure on/off numbers align with that notion.