Why PER Ain't no good

[Douglar]

Re: Advanced Stats, the Cavs, and Player Evaluation

I always use PER as a baseline, and combine with other stats. If you are a bigman, I look at REB% as well. If you are a PG, AST%. And always look at TS%.

I always also make sure to look at total minutes played. If you are looking at a PER for a period of less than 500 minutes, it get's really hard to tell what's noise and what's not.

 

[gourimoko]

Re: Advanced Stats, the Cavs, and Player Evaluation

"Bear in mind that this rating is not the final, once-and-for-all answer for a player's accomplishments during the season. This is especially true for players such as Bruce Bowen and Trenton Hassell who are defensive specialists but don't get many blocks or steals."

https://en.wikipedia.org/wiki/Player_efficiency_rating

I always use PER as a baseline, and combine with other stats. If you are a bigman, I look at REB% as well. If you are a PG, AST%. And always look at TS%.

Yeah I know, we've posted Hollinger's quotes before. He's on record that it's mostly useful as an offensive stat.. Dunno why this is getting lost.

 

[gourimoko]

Re: Advanced Stats, the Cavs, and Player Evaluation

No he has not. I've written this before and you ignored it, so let me make this as clear as possible: I CHALLENGE YOU TO FIND A STATEMENT FROM HOLLINGER TO THIS EFFECT.

I quoted Hollinger in the last thread.. So did someone else, twice.. And now a fourth person has right above..

Until you find it from Hollinger and quote it with a reference that the rest of us can see and confirm for ourselves, you don't have a leg to stand on. You are trying to fabricate your way out of a false claim that you made.

Dude you're being pretty obtuse when there's references coming at you left and right even from Wikipedia. Read your own thread man, I've cited him twice already as have others. Look two posts up! Hollinger's own words state that PER is not an all-inclusive metric, it mostly gauges offensive production. No one disputes this. No one.

I think you're entire argument is based on nothing and you haven't shown your work at all. There are several of us here with academic math backgrounds who are willing to entertain you, if you show your reasoning as to how "PER is mathematically flawed." However, you don't, you only cite others works which are only tangential to your claim and appeal to their academic backgrounds as an authority to your argument - which is lacking.

I've asked you for a proof that PER is "mathematically flawed," or does not correlate to an individuals offensive production. You haven't once demonstrated anything. You've cited another individual's work demonstrating that PER rewards inefficient chuckers. Everyone concedes this point. But you've made some wild claims that need to be backed up.

I think you ask for others to do so, and get angered when confronted with opposing evidence, because you don't have an argument. I'm wondering why you bothered to even make this thread, as the premise is ridiculous. PER works for what it's intended. It doesn't measure team statistics, and I don't think you understood that when you opened the thread.

 

[gourimoko]

Re: Advanced Stats, the Cavs, and Player Evaluation

No PER absolutely does not accurately rank the offensive talents of players.

Ugh..

Again:

Hollinger Stats - Player Efficiency Rating - Qualified Players
RK PLAYER GP MPG TS% AST TO USG ORR DRR REBR PER VA EWA
1 LeBron James, MIA 76 37.9 .640 23.3 9.6 28.2 4.4 20.8 13.1 31.67 909.0 30.3
2 Kevin Durant, OKC 81 38.5 .647 15.5 11.6 27.4 1.8 20.3 11.8 28.35 830.8 27.7
3 Chris Paul, LAC 70 33.4 .594 36.9 8.7 23.9 2.8 10.5 6.7 26.43 537.7 17.9
4 Carmelo Anthony, NY 67 37.0 .560 8.3 8.5 32.2 6.1 15.9 10.8 24.83 530.8 17.7
5 Brook Lopez, BKN 74 30.4 .567 4.8 8.9 26.1 10.8 16.1 13.4 24.81 477.8 15.9
6 Tim Duncan, SA 69 30.1 .554 12.8 10.2 24.8 7.3 29.6 19.1 24.45 401.7 13.4
7 Dwyane Wade, MIA 69 34.7 .571 19.3 10.6 27.0 4.7 12.6 8.9 24.04 483.1 16.1
8 Russell Westbrook, OKC 82 34.9 .532 22.8 10.2 31.2 4.8 11.8 8.6 23.98 554.4 18.5
9 Tony Parker, SA 66 32.9 .588 27.6 9.4 26.6 1.0 9.1 5.3 23.10 392.7 13.1
9 Kobe Bryant, LAL 78 38.6 .570 17.9 10.9 30.0 2.5 13.1 7.9 23.10 566.4 18.9
RK PLAYER GP MPG TS% AST TO USG ORR DRR REBR PER VA EWA
11 James Harden, HOU 78 38.3 .600 18.7 12.1 27.4 2.4 11.9 7.2 23.00 556.9 18.6
12 Blake Griffin, LAC 80 32.5 .572 17.2 10.6 24.0 8.7 21.5 15.2 22.44 424.3 14.1
13 Amar'e Stoudemire, NY 29 23.5 .637 3.4 12.9 22.8 10.0 15.1 12.4 22.16 108.5 3.6
14 Andray Blatche, BKN 82 19.0 .547 8.7 12.6 24.6 12.2 19.6 15.8 21.98 264.0 8.8
15 Anthony Davis, NO 64 28.8 .559 6.8 9.6 19.9 10.5 23.5 16.8 21.80 283.8 9.5
16 Anderson Varejao, CLE 25 36.0 .529 18.2 9.5 17.9 16.9 30.1 23.2 21.71 149.3 5.0
17 Andre Drummond, DET 60 20.7 .578 6.0 11.4 15.6 15.4 27.2 21.2 21.69 205.7 6.9
18 Kyrie Irving, CLE 59 34.7 .553 20.1 11.0 29.2 1.8 10.8 6.1 21.51 321.3 10.7
19 Stephen Curry, GS 78 38.2 .589 23.5 10.5 25.3 2.3 9.1 5.8 21.34 460.2 15.3
20 Brandan Wright, DAL 64 18.0 .606 7.3 6.6 16.8 8.4 16.7 12.6 21.03 178.9 6.0
RK PLAYER GP MPG TS% AST TO USG ORR DRR REBR PER VA EWA
21 Al Jefferson, UTAH 78 33.1 .522 10.2 6.6 23.4 7.0 25.9 16.3 20.99 399.8 13.3
22 John Wall, WSH 49 32.7 .521 26.7 11.2 28.7 2.4 11.2 6.8 20.91 236.9 7.9
23 JaVale McGee, DEN 79 18.1 .589 3.3 12.5 19.1 12.6 17.1 14.9 20.78 217.7 7.3
24 LaMarcus Aldridge, POR 74 37.7 .530 10.6 7.9 24.2 7.2 20.9 14.0 20.45 372.6 12.4
25 Deron Williams, BKN 78 36.4 .574 28.6 10.3 24.9 1.2 8.5 4.8 20.38 397.8 13.3
26 Nikola Pekovic, MIN 62 31.6 .572 5.1 9.7 20.3 13.0 19.0 15.9 20.26 282.3 9.4
27 DeMarcus Cousins, SAC 75 30.5 .524 12.1 13.7 25.9 10.9 27.0 18.7 20.21 328.3 10.9
28 David West, IND 73 33.4 .545 14.0 10.3 23.0 6.8 18.8 12.9 20.15 314.2 10.5
29 Chris Bosh, MIA 74 33.2 .592 9.5 9.9 20.0 7.0 17.6 12.6 20.08 347.3 11.6
30 Paul Millsap, UTAH 78 30.4 .550 14.7 10.1 21.1 8.6 18.8 13.7 19.89 297.4 9.9

Obviously there are some abberations, but the list correlates to what most people would consider the top-30 offensive talents in the league, give or take 5 players. It correctly identifies the top 4 players (imo) with 1-10 (minus Lopez) looking pretty accurate to me too.

Your quote:

No PER absolutely does not accurately rank the offensive talents of players.

Needs to be justified.

In fact, it's time to clearly addresses the even bigger issue with PER which so far not a single pro-PER person has mathematically discussed:

You have not demonstrated one single proof this entire thread. Now is the time to assert an argument? Not on the first page?

even as an offensive metric, PER is bad. It does NOT do a good job of pinpointing who is a good offensive player.

See chart above.. You're so wrong it's funny.

It routinely gives inflated PER numbers to players whose offensive play is preventing the team from winning because they are so terribly inefficient at scoring.

There is no perfect metric, I hope you're not claiming there is one. And can you demonstrate a metric more accurate at identifying the top ten offensive players in the league? Don't ignore this question, I'd like to see what you come up with.

That is the heart and soul of the critique of PER and not a single one of you pro-PER has said much about it. Curious.

Everyone has said tons about it. Myself, Tornicade, Douglar, and others have written pages that debunk your claims. The sources you cite don't even make the assertions you're making and you seem not to understand that the gap between you and your citations must be filled with an argument, analytically, logically or mathematically, yet you haven't demonstrated an argument at all. It's your thread.

 

[Douglar]

Meshuggener got very excited. He'll likely come back with in a week and a 1/2 from now with some insightful and polite responses. Right now he's on vacation.

 

[Douglar]

I thought this was an interesting list: Top NBA Centers

Which sorts more of the better players to the top? PER or WS?

 

[Pioneer10]

How in the world did Splitter get to be #4 in WS? He's above average but man he's soft as Charmin and he constantly messing up plays when I see him

 

[gourimoko]

I thought this was an interesting list: Top NBA Centers

Which sorts more of the better players to the top? PER or WS?

The list is very instructive. But of course, PER shows the top players, because that's what it's designed to do - calculate individual performance. Win Shares is very useful in determining the value a player has on his team, but not with respect to others. And before someone says "but Dwight Howard!?!," they should query the database again for career values, rather than only for last year.