Playground Tournament between Advanced Metrics?

28 Jan

Because I play pick up games and fantasy sports I often wonder…how would league/tournament look like if NBA players were distributed among teams by different metrics?

You can approach this topic in many ways but metrics agree about value of top stars so I don’t think standard draft of all players could work without issues typical to snake drafts in fantasy NBA.
Instead I wanted to focus and magnify differences in evaluation and here’s a twisted idea:
based only on last season’s numbers I subtracted player’s ranking according to given metric from average ranking for him based on 4 other statistics and sorted it by the biggest difference for each metric. In other words, I wanted to find players for each statistic which were valued the most by it when compared to others.

To make it clear, let me give you an example, last season Andrea Bargnani was ranked by PER as 80th best player in the league. Win Shares, Wins Produces, RAPM and ASPM graded him below Top200 [to be exact 216th, 245th, 231st, 209th for an average of 225.25].
So PER viewed Bargnani’s 2010-11 season 145.25 slots higher than the competition, and it was the highest difference for it, so he would be selected for the tournament as a representant for PER ;-) Got it?

The rest of such draft would look like this… [I included 9 top players for each metric with difference in parenthesis and extra players were mentioned because of positional need]

Team PER
Andrea Bargnani (145.25)
J.J. Hickson (132.5)
Brook Lopez (127.5)
Charlie Villanueva (113.25)
Antawn Jamison (111)
Richard Hamilton (110.5)
Andray Blatche (109.25)
Monta Ellis (107.5)
Al Jefferson (100.25)
Ramon Sessions – 11th

Team Win Shares
Zydrunas Ilgauskas (103)
Kurt Thomas (79.25)
Brandon Bass (77.75)
J.J. Redick (77.75)
Kyle Korver (76.75)
James Jones (75.75)
Serge Ibaka (73.75)
Zaza Pachulia (73.75)
Tyler Hansbrough (71.75)
Gary Neal – 24th
Louis Williams – 25th

Team Wins Produced
Landry Fields (119.5)
Ben Wallace (111.5)
Thabo Sefolosha (104.5)
Ed Davis (96.5)
Brandon Rush (96)
Kris Humphries (92.5)
Keith Bogans (89.75)
Derrick Favors (82.75)
Trevor Booker (80)
Jose Calderon – 18th

Ekpe Udoh (173)
Nick Collison (141)
Keyon Dooling (140)
Anthony Parker (139.75)
Donte Greene (137)
Vladimir Radmanovic (125.5)
Vince Carter (123.75)
Anthony Morrow (121.75)
C.J. Miles (117)
Anthony Tolliver – 19th
Jason Thompson – 21st
Roy Hibbert – 34th

Ben Wallace (84)
Monta Ellis (77.5)
Trevor Ariza (75.5)
Joel Anthony (74.75)
Kurt Thomas (70.5)
Amare Stoudemire (69)
Andris Biedrins (65.25)
Carlos Delfino (62.5)
Jason Richardson (59.75)
Mo Williams – 11th
Brandon Jennings – 12th

My first reaction to those teams was: “where are the point guards?”. 5 different point of views and yet neither of them contains a point guard in Top10 [unless you count tweener like Monta]
What does it mean? Do metrics mostly agree about point guard’s value?
Or simply do extremes present themselves in a best way among frontcourt players?
Also it creates two possible situations for the tournament, do we include point guards or play without them?

What’s more, those teams aren’t even close in terms of salaries!
According to data from 2010-11 team PER and team ASPM would be above salary cap, team RAPM would be around 55M$ and team WS and team WP would be below salary FLOOR.
Obviously it has something to do with rookie contracts but I think we should keep in mind those differences.

With all that said, here are some questions to consider just for fun…

1) Which matchup would you want to watch the most?

I would choose Team PER vs Team WP.
High-usage suspect-Defense vs Low-usage all-D? Count me in for that experiment ;-)

2) Which matchup would be the most one-sided?

I think team ASPM would destroy team RAPM mostly because the latter wouldn’t excel at any part of the game. Do you agree?

3) Which team would win the most games in a regular season for those 5 teams?

4) Which team would win the most games as a team in the NBA?

IMHO it would be team ASPM mostly because of the roster construction and complementary pieces.
But we should notice their differences for each player are just the least extreme of the bunch so ASPM basically accomplished being close to everybody ;-)

What do you think about all of this?


Posted by on January 28, 2012 in Unanswered Questions


Tags: , , , , ,

3 responses to “Playground Tournament between Advanced Metrics?

  1. Axe

    January 28, 2012 at 05:04

    “I think team ASPM would destroy team RAPM mostly because the latter wouldn’t be good at anything, right?” That’s certainly one way of saying “I have no idea what I’m writing blog posts about”

    • wiLQ

      January 28, 2012 at 15:11

      Or it was just a bad simplification… so thanks for pointing it out. I think it’s better now ;-)

  2. Alex

    January 29, 2012 at 04:47

    As a Pistons fan, I can tell you that having last year’s Charlie V and Rip Hamilton will get team PER nowhere.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s