PER vs. TPA vs. IQ

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • dell71
    Enter Sandman
    • Mar 2009
    • 23919

    PER vs. TPA vs. IQ

    Yay. Another NBA season is finally here.

    Let's look at who the best players were last year from a few different angles.

    Over the last few years, PER (Player Efficiency Rating) has become the gold standard in terms of basketball analytics. Like baseball's WAR (Wins Above Replacement), it has made its way onto our TV screens and into the conversations of the talking heads on those screens. It is a complicated formula that incorporates just about every box score stat an NBA player can get. It's a great, but flawed stat.

    I was curious to see how similar the results would be if we simplified things a bit. To do this I reached back for one of basketball's earliest metrics. I call it TPA (Total Production Average), but I'm sure there are other names for it. I've posted this for a past season once before, but I like to revisit this. What it does is tell us how "productive" a player is on a per-minute basis. A TPA of 0.850 or higher is basically All-Star level. It's extremely easy to calculate:

    ((Pts + Reb + Ast + St + Blk) - TOV)/Min

    Me being me, I got to tinkering around and came up with another stat all on my own that I'm tentatively calling IQ. The simple premise is to figure out how often a player does something "positive" (make a shot, grab a rebound, get an assist, etc) vs. how often that player does something negative (miss a shot, turn the ball over, commit a foul). This is a little more complicated but still easy to compute:

    A: (FGM + (3PM * .5) + FTM + Reb + Ast + St + Blk)
    (FGA + (3PA * .5) + FTA + Reb + Ast + St + Blk + TOV + PF)


    B: (lgFGM + (lg3PM * .5) + lgFTM + lgReb + lgAst + lgSt + lgBlk)
    (lgFGA + (lg3PA * .5) + lgFTA + lgReb + lgAst + lgSt + lgBlk + lgTOV + lgPF)


    C: (A/B) * 100

    That last part tells us how much better or worse than the NBA's average player is according to the formula. It sets the average player to 100 and provides the IQ. Therefore, a player with an IQ of less than 100 has a "low basketball IQ." I know, it doesn't really quantify a player's basketball IQ, just his penchant to contribute in a positive or negative manner, at least statistically. Hell, I may even change the name.

    Yup, both of the stats I've gone on about are flawed. Again, the object is to see how close (or not) I get to what PER gives us.

    What does PER give us?

    1. LeBron James-MIA 31.6
    2. Kevin Durant-OKC 28.3
    3. Chris Paul-LAC 26.4
    4. Carmelo Anthony-NYK 24.8
    5. Brook Lopez-BRK 24.7
    6. Tim Duncan-SAS 24.4
    7. Dwyane Wade-MIA 24.0
    8. Russell Westbrook-OKC 23.9
    9. Tony Parker-SAS 23.0
    10. Kobe Bryant-LAL 23.0
    11. James Harden-HOU 23.0
    12. Blake Griffin-LAC 22.4
    13. Andray Blatche-BRK 21.9
    14. Anthony Davis-NOH 21.7
    15. Kyrie Irving-CLE 21.4
    16. Stephen Curry-GSW 21.3
    17. Al Jefferson-UTA 20.9
    18. John Wall-WAS 20.8
    19. LaMarcus Aldridge-POR 20.4
    20. Deron Williams-BRK 20.3

    Okay, what about TPA?


    1. LeBron James-MIA 1.101
    2. Tim Duncan-SAS 1.051
    3. Kevin Durant-OKC 1.038
    4. Carmelo Anthony-NYK 0.992
    5. Russell Westbrook-OKC 0.991
    6. Kobe Bryant-LAL 0.956
    7. DeMarcus Cousins-SAC 0.945
    8. Brook Lopez-BRK 0.921
    9. Chris Paul-LAC 0.919
    10. James Harden-HOU 0.919
    11. Blake Griffin-LAC 0.910
    12. Al Jefferson-UTA 0.907
    13. Dwyane Wade-MIA 0.899
    14. Tony Parker-SAS 0.887
    15. John Wall-WAS 0.886
    16. Kyrie Irving-CLE 0.884
    17. Dwight Howard-LAL 0.880
    18. Andray Blatche-BRK 0.876
    19. LaMarcus Aldridge-POR 0.872
    20. David Lee-GSW 0.862

    We get a lot of the same names, but in a bit of a different order. Most notable: Tim Duncan, Russell Westbrook, DeMarcus Cousins, Dwight Howard, and David Lee shoot up the rankings (the latter three from outside the top 20) while Chris Paul, Andray Blatche, Anthony Davis, Stephen Curry, and Deron Williams take falls (those last three out of the top 20). King James still reigns, though.


    Now for IQ:

    1. Tyson Chandler-NYK 122
    2. Tim Duncan-SAS 117
    3. LeBron James-CLE 116
    4. Reggie Evans-BRK 116
    5. Joakim Noah-CHI 116
    6. JJ Hickson-POR 116
    7. Tiago Splitter-SAS 115
    8. Chris Paul-LAC 115
    9. Al Horford-ATL 114
    10. Nikola Vucevic-ORL 114
    11. Shawn Marion-DAL 114
    12. Kevin Durant-OKC 114
    13. Larry Sanders-MIL 113
    14. Serge Ibaka-OKC 113
    15. Al-Farouq Aminu-NOH 113
    16. Omer Asik-HOU 113
    17. Marc Gasol-MEM 113
    18. Andrei Kirilenko-MIN 113
    19. Marcin Gortat-PHO 113
    20. Anthony Davis-NOH 112
    Kenneth Faried-DEN 112

    Hmmm...now this is an interesting list. The first thing you'll notice is the name Tyson Chandler at the top. After you pick your jaw up from the floor and read down the list you'll see it's loaded with big guys who are not necessarily involved in the offense all that much. After all, it's hard to miss shots and/or turn the ball over if the rock doesn't come your way. This marks two guys as pretty amazing for cracking this list: LeBron James and Chris Paul. Unlike anyone else on this list, these two guys have the ball in their hands all the time, emphasis on all. Who takes shots on their teams is largely up to them. This is especially true of Paul who dictates what will happen on nearly every Clippers' possession when he's on the floor. Tim Duncan and Kevin Durant are also interesting entries given the rest of the list in that they are heavily involved in their team's offense. The difference between them vs. James and Paul is that these two guys aren't asked to facilitate. To me, IQ looks like a stat that has more value as a position specific tool than really judging anything. For instance, it might be helpful as a tool in grading point guards against one another but not a point guard against a center.

    Thoughts? Suggestions for improvements?


    Only players with 1500 or more minutes played were used.
Working...