Like us on Facebook


Follow us on Twitter





Results 1 to 5 of 5
  1. #1
    Join Date
    Jul 2013
    Location
    south
    Posts
    6,360
    vCash
    1500

    Question PIE (new NBA stat), What it measures, how good is it, how to find PIE info

    PIE, is the NBA's newest (?) stat. That's almost all I know about it. The NBA's Sekou Smith (MVP-Ladder, Hang-Time Blog) says it's the best stat ever. BUT, he does NOT explain either how it is compiled NOR why it is better than all the others. Additionally, while he does LIST it, he doesn't use it in his MVP-Ladder weekly column. Why?

    I posted this as part of a larger post in a different PSD "Advanced Stats" thread. There have been no responses and the thread wasn't really specifically about PIE. That's why I'm repeating my questions here.

    I sure wish for any help in understanding any and all aspects of it;
    INCLUDING, if possible, how to "produce" a PIE chart * of:
    The just-recently-completed regular season,
    ALL regular seasons,
    Playoffs and/or Finals.

    * Pun partially intended, hehe

  2. #2
    Join Date
    Nov 2009
    Location
    Bay Area
    Posts
    15,161
    vCash
    1500
    "Player Impact Estimate". It essentially takes all the box score stats we're used to seeing (points/assits/rebounds/blocks, etc), weights them how they feel appropriate (*0.5 added to Offensive boards and blocks for example) and kicks out a number for each game.

    In my opinion it's a really basic formula and in no way accounts for overall impact (specifically defensive impact) in a way that rivals a stat like RAPM. Like all the others, definitely not a great gauge for an "end all, be all" stat.

  3. #3
    Join Date
    Jul 2013
    Location
    south
    Posts
    6,360
    vCash
    1500
    Quote Originally Posted by tredigs View Post
    "Player Impact Estimate". It essentially takes all the box score stats we're used to seeing (points/assits/rebounds/blocks, etc), weights them how they feel appropriate (*0.5 added to Offensive boards and blocks for example) and kicks out a number for each game.

    In my opinion it's a really basic formula and in no way accounts for overall impact (specifically defensive impact) in a way that rivals a stat like RAPM. Like all the others, definitely not a great gauge for an "end all, be all" stat.
    Hey tredigs,
    Thanx for this.

    Is there someway/how accessible to the non-paying public to find the complete formula (plus any helpful explanations of the "why" for each "weighted" component?

    I tried really hard to use NBA.com's Stats "deparment"; but I never came close to being able to "produce" a PIE chart (neither for current playerS, nor for comparing players across the years). Have you had any success; or know of a "step-by-step" primer type thing to be able to do so?

    Less importantly, if you ever read Seku's "MVP Ladder", did you notice that despite his heavy praise for PIE, he never once used it as his deciding factor in his MVP weekly rankings?

    Do you think PER is better than PIE, about equal, or inferior. Are you saying that you value RAPM more than these and (all) other (advanced) stats?
    Last edited by Pablonovi; 05-20-2014 at 07:21 PM. Reason: "someone" -> "some way/how"; added last line/sentence

  4. #4
    Join Date
    Nov 2009
    Location
    Bay Area
    Posts
    15,161
    vCash
    1500
    Quote Originally Posted by Pablonovi View Post
    Hey tredigs,
    Thanx for this.

    Is there someway/how accessible to the non-paying public to find the complete formula (plus any helpful explanations of the "why" for each "weighted" component?

    I tried really hard to use NBA.com's Stats "deparment"; but I never came close to being able to "produce" a PIE chart (neither for current playerS, nor for comparing players across the years). Have you had any success; or know of a "step-by-step" primer type thing to be able to do so?

    Less importantly, if you ever read Seku's "MVP Ladder", did you notice that despite his heavy praise for PIE, he never once used it as his deciding factor in his MVP weekly rankings?

    Do you think PER is better than PIE, about equal, or inferior. Are you saying that you value RAPM more than these and (all) other (advanced) stats?
    No prob Pabs. Just go to Stats.NBA.com/glossary.html and actually click on each stat to have it open up and explained (and in the case of a formula, that is listed as well).

    Here is PIE's formula:

    (PTS + FGM + FTM - FGA - FTA + DREB + (.5 * OREB) + AST + STL + (.5 * BLK) - PF - TO) / (GmPTS + GmFGM + GmFTM - GmFGA - GmFTA + GmDREB + (.5 * GmOREB) + GmAST + GmSTL + (.5 * GmBLK) - GmPF - GmTO)

    Unfortunately I'm not ultra familiar with their stats department as far as making my own results or why they chose the exact formula they did for PIE here, but if/when I do I'll try to post my thoughts here.

    As far as RAPM, I think that (along with knowing the context of each player within their team as it pertains to them having a smaller/larger role) it gives some very "eye test confirming" results, especially on the defensive end. PER's very much an offensive skewed stat, as PIE seems to be also.

  5. #5
    Join Date
    Jul 2013
    Location
    south
    Posts
    6,360
    vCash
    1500
    Quote Originally Posted by tredigs View Post
    No prob Pabs. Just go to Stats.NBA.com/glossary.html and actually click on each stat to have it open up and explained (and in the case of a formula, that is listed as well).

    Here is PIE's formula:

    (PTS + FGM + FTM - FGA - FTA + DREB + (.5 * OREB) + AST + STL + (.5 * BLK) - PF - TO) / (GmPTS + GmFGM + GmFTM - GmFGA - GmFTA + GmDREB + (.5 * GmOREB) + GmAST + GmSTL + (.5 * GmBLK) - GmPF - GmTO)

    Unfortunately I'm not ultra familiar with their stats department as far as making my own results or why they chose the exact formula they did for PIE here, but if/when I do I'll try to post my thoughts here.

    As far as RAPM, I think that (along with knowing the context of each player within their team as it pertains to them having a smaller/larger role) it gives some very "eye test confirming" results, especially on the defensive end. PER's very much an offensive skewed stat, as PIE seems to be also.
    Hey tredigs,
    Thanx so much my man.
    It makes you wonder why the NBA itself can't do (much) better than PER - with all the resources at their disposal.

    I haven't "studied" RAPM much; absent that, I've long-favored the All-NBA Teams voting as the closest gage to what I thought I was seeing ON the court. I always thought they were far better (and more comprehensive than the MVP).

    The first time I felt strongly about this: When Bill Russell won the MVP two different times but didn't make first All-NBA Team Center (deservedly NOT, in my opinion, in either of those years).

    In any event, I DO think the composite voting of the All-NBA TeamS (plural) comes pretty darned close to being a very accurate summation of Regular Season play.

    You take?

    Oh, I'm AM looking forward to your (further) thoughts on PIE.

    There was ONE time when I got the NBA Stats section to come up with a complicated chart I really liked; but then I couldn't copy it - to be able to further work with the info (is that on purpose that you can't copy it; or is it just my lack of net-skills???). It seems to me they could/should make it easier on us net-weaklings!

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •