Offensive All Stars

Posted on 01/07/2011 by

17



“Before the beginning of great brilliance, there must be chaos. Before a brilliant person begins something great, they must look foolish in the crowd.”-I Ching

Sometimes I feel like my work is never done. I do one thing and rather than an ending it just leads to a new beginning.  I come up with something that I think is just so only to find I am wrong or that there are miles still to go to achieve that magical state of being done. I guess it’s because I forget that the gaining of  knowledge and wisdom is a processes and a journey and not a simple destination.

Here's to the damn process (Image courtesy of xkcd.com)

As you know if you read this blog regularly, I’ve been working on a simple metric to quantify offensive efficiency. I’ve actually done a whole series on this if you’re interested:

Gimme the Rock The Series:

The Intro and the Paradox of Melo

Point Guards and the Nash Equilibrium

Shooting Guards and the Shooters Dilemma

Small Forwards: Jacks of all Trades

Spreading the wealth

The goal is to develop a simple measure for the return (points) on investment (possessions) for players on offense (i.e. Player offensive efficiency). The concept is based on the idea that were I an NBA gm paying a player I would care about getting value (points) from my assets (possessions). I want to do this simply in a way that anyone can understand and using publicly available information.

I though the last version of the metric was just about there. Then, as I noted in the shot creationpost, I read this comment:

Austin,

Arturo, there’s one thing I think is essential to improving this ranking: individual modifiers for number of assisted field goals. Using stats from hoopdata:
http://hoopdata.com/scoringstats.aspx

the %ast column gives the ratio of assisted FG to unassisted, and especially for high-usage players, this number is often much lower than 70% (and varies greatly). So I think that’s an important change that will shake up the rankings quite a bit.

Great stuff as always.

Damn Austin  for being totally right . So I went back to the drawing board somewhat and I dug in at hoopdata.  I ended up adding %Assists (Percent of Field Goals from Assists) and % of Assist leading to 3PM to the calculation. So the detail on the metric is now as follows:

Look at all the data for the Current Season for every player and get:

  • Player Position
  • Minutes Played (and eliminate all players with less than 180 Minutes Played this year which leaves 332 players)
  • Pts48 (points per 48 minutes)
  • FGA48 (field goal attempts per 48 minutes)
  • Pts per Attempt (pts per attempt = pts48/fga48)
  • AST48 (assists per 48 minutes)
  • TO48 (turnovers per 48 minutes)
  • FTA48. This was added because of two of my fabulous readers (some dude and Man of Steele take a bow) who pointed out that some free throws while free do end possession. They’re right of course. So we are going to add the SD-MOS hack-a-shaq correction.
  • %Assists (Percent of Field Goals from Assists)
  • % of Assist leading to 3PM

Then I get to work making calculations:

  • Offensive possessions used per 48 (FGA48+AST48+TO48 +.44 *FTA48 term (an approximation for possessions used thru Free Throws from Prof. Berri) this is the number of possesions spent by the player)
  • Multiplier for Points per 48 or Point Value: 1-Asst%/2 . This is the first part of making sure that the value of the Assisted field goals gets split between the passer and the shooter.
  • Assist Value: 1+ % of Assists leading to 3Point Field Goals/2. This is the second part of making sure that the value of assisted field goals get split between the passer and the shooter (and that we account for all three pointers)
  • Offense Generated per 48 Minutes: Pts48*(Point Value for each player)  +AssistsP48 *(AssistValue) . So no double counting on assists, all the splits are done by player and 3PM are properly accredited. Yay.
  • Offense generated per possession used. This is the key measure as it reflects how many points the team generates when the player in question gets the rock. This is OffGenP48/ OffPossUsedP48

  • Offense Generated at 30 possessions used. Here I’m just projecting every player at an even number of possessions. So Offense per possesion times 30.

Finally we group everyone by position and usage quartiles and this gives us the Gimme the Rock Metric take 3 or as it will be now know the Offensive Ranking.

Offensive Ranking for the 2010-2011 Season so far:

But that’s a lot of data. Let’s look at the above average performers. Who are the offensive All-Stars?

Player Team Position Usage Group by position Offense Generated at 30 possesions used Standard Deviations from Mean
Steve Nash PHO 1 1VeryHigh 30.2 2.5
Chris Paul NOH 1 1VeryHigh 30.1 2.4
Luke Ridnour MIN 1 3Low 29.3 2.0
Jameer Nelson ORL 1 2High 28.0 1.5
Tony Parker SAS 1 1VeryHigh 27.7 1.3
Ty Lawson DEN 1 3Low 27.7 1.3
Ramon Sessions CLE 2 1VeryHigh 26.2 1.9
Dwyane Wade MIA 2 1VeryHigh 26.2 1.9
Kirk Hinrich WAS 2 3Low 25.8 1.7
Manu Ginobili SAS 2 1VeryHigh 25.6 1.6
Jamal Crawford ATL 2 1VeryHigh 24.9 1.3
Monta Ellis GSW 2 1VeryHigh 24.7 1.3
Tracy McGrady DET 2 3Low 24.6 1.2
J.J. Redick ORL 2 3Low 24.5 1.2
Kevin Martin HOU 2 1VeryHigh 24.4 1.2
Kobe Bryant LAL 2 1VeryHigh 24.3 1.1
Daniel Gibson CLE 2 2High 24.2 1.1
Eric Gordon LAC 2 1VeryHigh 24.2 1.1
LeBron James MIA 3 1VeryHigh 26.6 2.7
Paul Pierce BOS 3 1VeryHigh 25.9 2.4
Rudy Gay MEM 3 1VeryHigh 24.7 1.9
Grant Hill PHO 3 2High 24.3 1.7
Brandon Roy POR 3 1VeryHigh 24.2 1.6
Tayshaun Prince DET 3 2High 24.1 1.6
Josh Childress PHO 3 4VeryLow 23.2 1.2
Carmelo Anthony DEN 3 1VeryHigh 23.2 1.2
Reggie Williams GSW 3 2High 23.1 1.1
Andre Iguodala PHI 3 2High 23.0 1.1
Danilo Gallinari NYK 3 3Low 22.9 1.1
Shawne Williams NYK 4 4VeryLow 26.2 2.3
Lamar Odom LAL 4 2High 24.6 1.7
Dirk Nowitzki DAL 4 1VeryHigh 24.3 1.6
Craig Smith LAC 4 2High 24.0 1.5
Amir Johnson TOR 4 3Low 23.7 1.3
Chris Bosh MIA 4 1VeryHigh 23.3 1.2
Hedo Turkoglu TOT 4 2High 23.3 1.2
Chuck Hayes HOU 4 4VeryLow 23.1 1.1
Wilson Chandler NYK 4 2High 23.1 1.1
David West NOH 4 1VeryHigh 23.0 1.1
Nene Hilario DEN 5 2High 25.1 2.1
Pau Gasol LAL 5 1VeryHigh 24.4 1.8
Al Horford ATL 5 1VeryHigh 23.8 1.6
Amare Stoudemire NYK 5 1VeryHigh 23.5 1.5
Ronny Turiaf NYK 5 4VeryLow 23.4 1.4
Brad Miller HOU 5 1VeryHigh 23.3 1.4
Dwight Howard ORL 5 1VeryHigh 23.0 1.3
Joakim Noah CHI 5 2High 22.9 1.2
Tyson Chandler DAL 5 4VeryLow 22.7 1.1
DeAndre Jordan LAC 5 4VeryLow 22.5 1.1
Marc Gasol MEM 5 2High 22.4 1.1
Jon Brockman MIL 5 4VeryLow 22.3 1.0
Melvin Ely DEN 5 4VeryLow 22.3 1.0

We have 52 players and it breaks down as 6 PG,12 SG, 11 SF, 10 PF and 13 Center.The super outstanding (2 std deviation above the norm for their position) are Nash,Paul and Ridnour at PG, no SG (I’ve said before that the value here is not in shooting), Lebron and Pierce at the 3, Shawnee Williams at the 4 (this bears investigating and may be a function of small sample size, that said if I’m Denver I might inquire as to his availability as a throw in in a Carmelo deal), and Nene alone at the 5.

The answer for the most outstanding is the same answer that seems t0 always come up:

Hail to the King again

Here’s to evolution.

Awesome

Advertisements
Posted in: Uncategorized