Every Little Bit Helps – Rim Protection From All Angles

Jan 16, 2015; Philadelphia, PA, USA; New Orleans Pelicans forward Dante Cunningham (44) drives the basket and collides with Philadelphia 76ers guard K.J. McDaniels (14) during the second half at Wells Fargo Center. The 76ers defeated the Pelicans 96-81. Mandatory Credit: Bill Streicher-USA TODAY Sports
Jan 16, 2015; Philadelphia, PA, USA; New Orleans Pelicans forward Dante Cunningham (44) drives the basket and collides with Philadelphia 76ers guard K.J. McDaniels (14) during the second half at Wells Fargo Center. The 76ers defeated the Pelicans 96-81. Mandatory Credit: Bill Streicher-USA TODAY Sports /
facebooktwitterreddit
Jan 16, 2015; Philadelphia, PA, USA; New Orleans Pelicans forward Dante Cunningham (44) drives the basket and collides with Philadelphia 76ers guard K.J. McDaniels (14) during the second half at Wells Fargo Center. The 76ers defeated the Pelicans 96-81. Mandatory Credit: Bill Streicher-USA TODAY Sports
Jan 16, 2015; Philadelphia, PA, USA; New Orleans Pelicans forward Dante Cunningham (44) drives the basket and collides with Philadelphia 76ers guard K.J. McDaniels (14) during the second half at Wells Fargo Center. The 76ers defeated the Pelicans 96-81. Mandatory Credit: Bill Streicher-USA TODAY Sports /

Based on a question from Kelly Scaletta, I wanted to investigate how much value the shot blocking and/or rim protection[1. Not at all the same thing in many cases.] of a particularly active wing player like K.J. McDaniels might provide to a team. In the past, I’ve looked into rim protection for players at all positions, but have generally only presented the results for power forwards and centers, both to avoid clutter and because I’m more confident in those results.  For players operating more on the perimeter, any such numbers are probably heavily influenced by if the player tends to get help from a secondary shot blocker or not.

Again, blocked shots aren’t the end-all, be-all but according to SportVU shot logs, 3.1% of shots where a PF or center was the primary defender were blocked by other (non-primary defender) players, but over 4.8% of similar shots where the primary defender was a PG or wing were blocked by help defenders. Still, some numbers for wings and points are better than none.

Before simply dumping a bunch of charts, a quick refresher about the methodology. The basic analysis is on two levels. First, since we know the league average FG% for uncontested shots at or near the rim as well as the average allowed by each player per SportVU, the difference between points allowed and what would be expected on those same shots with no defense is sort of a “raw” points saved. Of course, basing the analysis on what would have happened should the player simply have been removed from the court, leaving the defense playing with 4 guys[2. Insert Kings jokes here.] is unrealistic. So the second part of the examination is finding what a theoretical “average” defender would  have saved and comparing the two.

To ensure it is apples-to-apples, the comparison is to a positional average.  As position in the NBA is often defined by “who are you guarding?” I think this captures the different expectations and rolls to a degree, as one big man provides as much rim protection on average as all three perimeter players combined per minute. [3. There is also a slight adjustment involved to account for players playing at a higher or lower pace and for teams allowing more or fewer close shots. Essentially I wanted to avoid over-crediting players who happened to contest a lot of shots because their teams were playing poor enough defense to allow defenders many more “opportunities” to defend the rim.]

I also modified the formulas slightly because I wasn’t happy with how the model was dealing with players slightly above or slightly below average in terms of percentages allowed, as I think I was overweighting number of contests at the the expense of percentage allowed.[3. This modification proved to be a reminder of just how bad my R skills were when I first started, as I had to rewrite the whole thing more or less from scratch. And I’m now terrified to go back through the code of some of my other early metrics to see how ugly it really is.] So if you’ve been following this stat for a while, you might notice some slight adjustments in the ordering of players as well as a slightly larger spread from best to worst among big men.

And as a final reminder, rim protection is only a portion of defense. My estimate, backed up by studies from Kevin Hetrick and Justin Willard suggests rim protection is about 40% of the defensive contribution for a big man, probably substantially less for wings and points, especially considering, as the charts below show, the range between best and worst wings and point guards is substantially smaller than that found between the most and least effective big men.

That said, a player like McDaniels could indeed provide some extra value with his ability to defend the rim from unusual spots. Certainly it’s no surprise that top wing defenders like Danny Green and Michael Kidd-Gilchrist also show up well in terms of this additional value. Again, all numbers are measured against a hypothetical average defender from the relevant position, point, wing or big.

Bigs:

Wings:

Point Guards: