Pages Navigation Menu

Kevin Durant Kills it from Everywhere: Introducing Scoring Versatility Index

US Presswire

US Presswire


 
There are a number of different stats to quantify and qualify scoring prowess in the NBA. Generally they fall into either volume statistics or efficiency stats with points per game and true shooting being the kings of their respective categories. In a Grantland article last week Kirk Goldsberry unveiled a new shooting statistic, ‘Shot Score.’ Essentially shot score is the inverse of Ian Levy’s XPPS, the expected points per shot given a player’s overall shot selection. Instead, Goldsberry emphasizes the differential between the player’s expected point production and actual point production, treating it something like a degree of difficulty in diving.

Goldsberry explains that someone like LeBron James who scores efficiently from a variety of places on the court is more valuable than a player like DeAndre Jordan even though Jordan has a higher true shooting percentage, the traditional advanced scoring stat, than James.

I think that is at least partly true, the ability to score efficiently from a variety of spots on the floor is likely to make it much harder to guard a player and allow the offense more flexibility. To some degree this is realized in the scoring differential between Jordan and James. Because DeAndre Jordan can only score effectively in a limited amount of real estate he got half the shots per 48 minutes and got to the line for three fewer attempts per 48 minutes of play than James according to Hoopdata, even with his poor free throw shooting inviting Hack-a-DeAndre at times.

However, I noticed that the way shot score was calculated, adding up the net points a player generated over the expected points, could reward a player like DeAndre Jordan or Steve Novak.  Goldsberry acknowledged as much in a podcast with Levy on Hickory, pointing to his earlier work at the Sloan Sports Conference.

In fact, it was probably the disproportionate efficiency James showed inside that induced the Spurs to adopt the defensive strategy they came one free throw or offensive rebound away from riding to a Finals victory. Instead of blanketing James with tight coverage, or bring a double team on James, who is an excellent passer out of the double, the Spurs deliberately gave both James and his team mate Dwayne Wade space to shoot with only a loose contest daring them to pull up and shoot. (Also, taking advantage of the fact that it is generally harder to, shoot off the dribble).

I viewed the Spurs strategy as another demonstration of the value of shooting versatility, (as well as, just how effective James is at scoring in the paint). I also thought it would be very unlikely such a strategy would be successful against Durant, who is unusually lethal from just about all over the court.

So, I decided to try to create a metric that measured scoring versatility in the NBA, which is different from how I interpret the Goldsberry measure as shooting prowess adjusted for degree of difficulty. Instead, I wanted to create a measure that did track how versatile players were in their scoring, including the ability to get to the line as Tom Ziller at SB Nation emphasized.

To build the index I used Hoopdata.com’s shot data, that I have used in the past on shot locations, along with their free throw data. I divided each player’s efficiency at HoopData’s five shot locations by the average player’s efficiency. The five locations are, At the Rim, Short (three to nine feet), Mid-Range (10-15 ft), Long-Two’s (16-23 ft) and Threes, along with made free throws per field goal attempt. Then I took the average of those scores times the percentage of zone’s where the player attempted a shot.

The intent was to measure how many different locations a player could score efficiently. Like any efficiency measure there were sample size issues, but limiting the selection of players with at least 250 shot attempts gave me very presentable results.

Below are the top twenty players in scoring versatility, the higher the ratio above one the more efficient the player was compared to the average player:

Name Rim Ratio Short Ratio
(3-9)
Mid Ratio
(10-15)
Long Ratio
(16-23)
Three Point
Ratio
FTM/FGA
Ratio
Versalitiy Index
Kevin Durant 1.21 1.55 1.49 1.26 1.47 2.49 1.51
LeBron James 1.26 1.72 1.08 1.35 1.44 1.57 1.33
Chris Paul 1.12 1.58 1.33 1.47 1.16 1.76 1.32
Dirk Nowitzki 1.02 1.48 1.42 1.47 1.48 1.20 1.27
Kobe Bryant 1.12 1.40 1.37 1.17 1.16 1.73 1.27
James Harden 1.01 1.08 0.77 1.06 1.30 2.65 1.25
Steve Nash 1.14 1.21 1.34 1.47 1.55 1.19 1.24
Tony Parker 1.11 1.43 1.22 1.29 1.23 1.48 1.23
Carl Landry 1.14 1.10 1.21 1.11 1.18 1.95 1.23
Deron Williams 1.13 1.28 1.27 1.17 1.34 1.48 1.22
Serge Ibaka 1.25 1.35 1.54 1.38 1.25 0.97 1.22
Darren Collison 1.08 1.19 1.15 1.23 1.27 1.76 1.21
Jason Terry 0.95 1.63 1.37 1.35 1.32 0.93 1.19
Amir Johnson 1.07 1.60 1.17 1.09 1.37 1.18 1.19
Ray Allen 0.95 1.68 1.21 0.97 1.50 1.13 1.19
Kevin Martin 1.19 0.78 1.30 1.26 1.52 1.48 1.19
Kyrie Irving 0.95 1.30 1.31 1.32 1.39 1.22 1.18
Jamal Crawford 1.08 1.56 1.20 1.14 1.33 1.12 1.18
Stephen Curry 0.95 1.18 1.42 1.29 1.61 0.99 1.17

Kevin Durant and LeBron James were both masterful scorers, but Durant’s superior Mid-Range game and the ability to get the line pulled him ahead.  James Harden wouldn’t even make the list without including free throws.  Also of note, Carl Landry, who is likely to miss the season, was surprisingly versatile in his game last year; above average from everywhere.

On the other end, here are the low scorers in terms of scoring versatility.  The bottom of the list is dominated by front court players, not surprisingly, and in many cases may not be an issue for their team’s offense unless the lack of any mid-range game clogs the team’s offensive spacing.

Name Rim Ratio Short Ratio
(3-9)
Mid Ratio
(10-15)
Long Ratio
(16-23)
Three Point
Ratio
FTM/FGA
Ratio
Scoring Zone
Coverage
Versalitiy Index
Josh Childress 0.89 0.00 0.00 0.00 1.18 0.25 0.67 0.26
Bismack Biyombo 0.98 0.75 0.88 0.61 0.00 1.13 0.83 0.58
Lavoy Allen 1.08 1.31 0.80 1.00 0.00 0.50 0.83 0.61
Kosta Koufos 1.07 1.40 0.88 0.73 0.00 0.60 0.83 0.62
Kevin Seraphin 1.12 1.37 1.02 1.02 0.00 0.45 0.83 0.65
Larry Sanders 1.05 0.75 0.80 0.85 0.00 0.73 1.00 0.65
Emeka Okafor 1.13 1.23 1.03 0.85 0.00 0.84 0.83 0.67
Reggie Evans 0.86 0.93 0.00 1.23 0.00 2.28 0.83 0.68
Kenneth Faried 1.07 1.14 0.88 0.91 0.00 1.25 0.83 0.69
Jason Maxiell 1.10 1.40 0.78 0.94 0.00 1.04 0.83 0.69
Elton Brand 0.97 1.40 1.03 1.26 0.00 0.73 0.83 0.69
John Henson 1.06 1.14 0.60 0.73 0.00 0.88 1.00 0.70
DeAndre Jordan 1.18 1.55 1.05 0.38 0.00 1.04 0.83 0.71
Nikola Pekovic 1.01 0.99 0.97 0.67 0.00 1.71 0.83 0.71
Lamar Odom 0.96 0.99 0.58 1.14 0.71 0.29 1.00 0.72
Charlie Villanueva 0.97 1.18 0.29 0.50 1.23 0.31 1.00 0.72
Keith Bogans 1.21 1.71 0.00 1.35 1.20 0.20 0.83 0.73
Omer Asik 0.97 0.77 1.44 1.02 0.00 1.44 0.83 0.74
Taj Gibson 1.11 0.62 1.03 0.94 0.00 1.06 1.00 0.74
Chris Kaman 1.03 1.64 1.12 1.49 0.00 0.57 0.83 0.75

Bottom line, I think the degree to which the ability of a player to score from a variety of spots on the floor is more valuable than a specialist is very much an open question.  Efficient use of possessions is the ultimate goal and, to me, the starting place for any analysis.  But, given that every team is trying to score against a live team defense trying to deny the offense open shots from their preferred locations it is probably helpful to make them worry about denying as many spots as possible.

  • john

    Actually what I read into the data is that Harden takes few of the worst shot in basketball. Unless a player is a marksman like Durant, statistically every one of those mid range shots actually hurts.

    • Andrew Johnson

      On Harden that’s true, Ian did some great work on that with his Xpps. But what really makes Harden elite is his ability to get the line. He had a 50.8% eFG, but a 60 TS%. Both he and Durant came out as big outliers in a free throw shooting I did.

      • john

        Thanks. And I do love your articles.

%d bloggers like this: