PER vs. Game Score vs. NBA Efficiency

Moderators: Harry Garris, ken6199, Dirk, bisme37, KingDavid, bwgood77, zimpy27, cupcakesnake, Domejandro, infinite11285

Minge
Bench Warmer
Posts: 1,421
And1: 6
Joined: Jul 03, 2006

Re: PER vs. Game Score vs. NBA Efficiency 

Post#21 » by Minge » Fri Aug 14, 2009 9:20 pm

What stat are you talking about? If its stats in general then your being way too vague. If I told you so and so was a 80% FT shooter, would you counter with well thats not a true measure of his ability?

That comment was in regard to "PER". Interestingly, Rick Barry has claimed that free throw shooting percentage is the only "trustworthy" statistic in basketball.

"Before delving into what happened in this game, I have to mention a scorekeeping issue that really bothers me--and a problem that Hall of Famers Oscar Robertson and Rick Barry have talked about for years. Robertson, at one time the career leader in assists, has repeatedly said that assists are doled out much more generously by scorekeepers than they were when he played. Barry told me that the only statistic he trusts is free throw percentage because every other number can be manipulated in some way either by the player (for instance, missing a shot to pad offensive rebounding totals) or by the subjective judgment of the scorekeeper, who has the final word in deciding whether or not a tap is an offensive rebound, who should get credit for a steal--and what constitutes an assist." - May 20, 2008
Don Draper
General Manager
Posts: 8,677
And1: 506
Joined: Mar 09, 2008
Location: schönes Wetter

Re: PER vs. Game Score vs. NBA Efficiency 

Post#22 » by Don Draper » Fri Aug 14, 2009 9:25 pm

BrooklynBulls wrote:Using Dave Berri as a source is a lot like using Wikipedia as a source, but without the credibility. The guy...is a moron with a knowledge of advanced statistical theory. So he knows a lot about math, and almost nothing about basketball. Consult his wages of win statistic for further evidence.

He's a moron and knows nothing about basketball yet he is an NBA consultant? :roll:

I'm not a big fan of Dr. Berri but he writes interesting articles. Although I am unsure how valid his wins produced metric don't you find it odd that there is a .99 correlation between PER and other metrics other people disregard? When it comes down to it, PER measures the exact same things with the exact same accuracy.

Though you may not think his wins produced passes the laugh test why is it that adjusted +/- yields similar questionable results and NBA teams still use it (both use linear regression). Furthermore, unless you read the book I doubt you cite specific examples where it doesn't make sense beside how you feel the players should be ranked.

FYI PER also uses weights.
soda wrote:I will never, ever, ever vote for a socialist. I'd vote for a member of the KKK first. I'd vote for Hitler first, because the Nazis have less blood on their hands

This is the state of modern day political discourse.
User avatar
BrooklynBulls
Retired Mod
Retired Mod
Posts: 32,733
And1: 2,652
Joined: May 13, 2007
Location: Avidly reading WillPenney.com
Contact:

Re: PER vs. Game Score vs. NBA Efficiency 

Post#23 » by BrooklynBulls » Fri Aug 14, 2009 11:02 pm

obinna wrote:
don't you find it odd that there is a .99 correlation between PER and other metrics other people disregard? When it comes down to it, PER measures the exact same things with the exact same accuracy.


I find it interesting. I would like to see the regression that produces the correlation, but I didn't see it in the article.

Though you may not think his wins produced passes the laugh test why is it that adjusted +/- yields similar questionable results

I've not seen that data, although I'd argue that +/- does not have too much to do with efficiency, but how well a player compliments a team's needs. Since wages of win is attempting to quantify player production, and not fit, I think that this effect occurs because Berri designed his statistical model to adhere to this correlation. With that said, I'd like to see the data.

Furthermore, unless you read the book I doubt you cite specific examples where it doesn't make sense beside how you feel the players should be ranked.


It doesn't make sense because it's not weighted, and as a result, values a rebound as much as a point. But because points are penalized because misses are a result of attempting to score, yet rebounds are NOT penalized because there is no way to miss a rebound, the result is that mediocre big men dominate talented guards, and in my basketball opinion, that's trash.

FYI PER also uses weights.


Yeah, I know. I have mixed feelings about that- arbitrary, statistically regressed weights have statistical evidence behind them, but there is a basketball purist in me that feels that because stats are naturally confounded- several stats happen simultaneously, and one stat leads to another, these variables cannot be separated and weighed like so much hay. But, though an imperfect system, PER passes the eye test for me, and I understand its' derivation and its' usefulness. Because it levels the playing field, I prefer it to many other advanced stats-- it's just used out of context a lot.
Chronz
Starter
Posts: 2,199
And1: 467
Joined: Jul 30, 2008

Re: PER vs. Game Score vs. NBA Efficiency 

Post#24 » by Chronz » Fri Aug 14, 2009 11:12 pm

Minge wrote:
What stat are you talking about? If its stats in general then your being way too vague. If I told you so and so was a 80% FT shooter, would you counter with well thats not a true measure of his ability?

That comment was in regard to "PER". Interestingly, Rick Barry has claimed that free throw shooting percentage is the only "trustworthy" statistic in basketball.

"Before delving into what happened in this game, I have to mention a scorekeeping issue that really bothers me--and a problem that Hall of Famers Oscar Robertson and Rick Barry have talked about for years. Robertson, at one time the career leader in assists, has repeatedly said that assists are doled out much more generously by scorekeepers than they were when he played. Barry told me that the only statistic he trusts is free throw percentage because every other number can be manipulated in some way either by the player (for instance, missing a shot to pad offensive rebounding totals) or by the subjective judgment of the scorekeeper, who has the final word in deciding whether or not a tap is an offensive rebound, who should get credit for a steal--and what constitutes an assist." - May 20, 2008

Believe me if I could I would counter to Rick Barry and Big O. Facts are you can measure rebounds coralled from a players own miss but it is neither stat padding nor relevant. For example a player missing the shot and rebounding his own miss would result in a net score of 0PTS gained 0Efficiency lost, on any VORP scale it wouldnt impact a players worth one iota. Assist are by far the most subjective stat so I agree with him there, but again you can measure the influence the definitional change had on league averages. Assists in Big O's era are valued more than they are in an era with a 20% increase or whatever it was, and steals are usually credited to the right players, but yes in times where it couldnt have happened without the participation of both players Id like to see .5 stls handed out. I dont know why the NBA insists on tracking such useless stats and ignoring the important ones, then again they didnt differentiate between rebounds either.

The good news is that the NBA is going to get awhole lot more technological with their stats, should shed some light on defenses.
Patterns
Banned User
Posts: 6,008
And1: 18
Joined: Sep 19, 2007

Re: PER vs. Game Score vs. NBA Efficiency 

Post#25 » by Patterns » Fri Aug 14, 2009 11:47 pm

twenty23 wrote:
JimMurray wrote:Wow, it really only took three posts to turn this into Kobe vs. Lebron?

Legends are measured in rings. Kobe 4, Lebron 0.

Thats why Bill Russel will always be regarded as being >>>>>>> than Wilt Chamberlain.


Rob Horry > MJ

FACT.

Everytime someone points out that rings are more important than anything else, someone always brings up Robert Horry.

It's almost like Horry single handily made winning rings pointless to these fans of ringless superstars.
User avatar
Magic Mamba
Junior
Posts: 434
And1: 22
Joined: Jun 22, 2008

Re: PER vs. Game Score vs. NBA Efficiency 

Post#26 » by Magic Mamba » Fri Aug 14, 2009 11:53 pm

twenty23 wrote:Kobe homers are scared of per.



Just like LeBITCH is scared of Championship Rings. :lol:

Image

Image
Image
User avatar
No Offense
Assistant Coach
Posts: 4,202
And1: 260
Joined: Aug 28, 2008

Re: PER vs. Game Score vs. NBA Efficiency 

Post#27 » by No Offense » Fri Aug 14, 2009 11:58 pm

doozyj wrote:
LebronsCavs wrote:
doozyj wrote:PER is used to make some RealGM members favorite players to appear better than they really are.


:lol: Lebron >>>Kobe. Deal with it


Sure whatever you say. Lebron can keep being better all he wants as he runs off the court, crying as he continues to lose all while having no rings. :roll:

Lebron is overrated by the Media and by his padded stats. Deal with that.


Says the troll.
The Skyhook
RealGM
Posts: 11,432
And1: 925
Joined: Sep 16, 2008
 

Re: PER vs. Game Score vs. NBA Efficiency 

Post#28 » by The Skyhook » Sat Aug 15, 2009 5:55 am

All it takes is one idiot to turn this into a Kobe/Lebron debate.

Just STFU they're both good. Better than everyone else who plays the game today.
jeahwe
Junior
Posts: 436
And1: 1
Joined: Aug 08, 2006

Re: PER vs. Game Score vs. NBA Efficiency 

Post#29 » by jeahwe » Sat Aug 15, 2009 8:58 am

BrooklynBulls wrote:Using Dave Berri as a source is a lot like using Wikipedia as a source, but without the credibility. The guy...is a moron with a knowledge of advanced statistical theory. So he knows a lot about math, and almost nothing about basketball. Consult his wages of win statistic for further evidence.


You couldn't be more wrong.
Berri's metric is the best thing avaliable.

http://dberri.wordpress.com/2006/11/17/ ... cy-rating/

Looking at the specific weights Hollinger chooses we see another problem. In discussing the NBA Efficiency metric – which the NBA presents at its website – I argued that this measure fails to penalize inefficient shooting. The regression of wins on offensive and defensive efficiency reveals that shooting efficiency impacts outcomes in basketball. The ball does indeed have to go through the hoop for a team to be successful.

The same critique offered for NBA Efficiency also applies to Hollinger’s PERs, except the problem is even worse. Hollinger argues that each two point field goal made is worth about 1.65 points. A three point field goal made is worth 2.65 points. A missed field goal, though, costs a team 0.72 points.

Given these values, with a bit of math we can show that a player will break even on his two point field goal attempts if he hits on 30.4% of these shots. On three pointers the break-even point is 21.4%. If a player exceeds these thresholds, and virtually every NBA played does so with respect to two-point shots, the more he shoots the higher his value in PERs. So a player can be an inefficient scorer and simply inflate his value by taking a large number of shots.
jeahwe
Junior
Posts: 436
And1: 1
Joined: Aug 08, 2006

Re: PER vs. Game Score vs. NBA Efficiency 

Post#30 » by jeahwe » Sat Aug 15, 2009 9:04 am

BrooklynBulls wrote:I find it interesting. I would like to see the regression that produces the correlation, but I didn't see it in the article.


Read Berri's book.
Some things you could also find here: http://www.wagesofwins.com/CalculatingWinsProduced.html
User avatar
dflash3
Retired Mod
Retired Mod
Posts: 10,733
And1: 157
Joined: Dec 08, 2005
 

Re: PER vs. Game Score vs. NBA Efficiency 

Post#31 » by dflash3 » Sat Aug 15, 2009 9:35 am

twenty23 wrote:Kobe homers are scared of per.

LOL that was the first thing that entered my mind before I opened the thread and wouldn't you know the very first comment from a Laker fan confirmed it. :lol:
Image
mysticbb
Banned User
Posts: 8,205
And1: 713
Joined: May 28, 2007
Contact:
   

Re: PER vs. Game Score vs. NBA Efficiency 

Post#32 » by mysticbb » Sat Aug 15, 2009 11:09 am

jeahwe wrote:http://dberri.wordpress.com/2006/11/17/ ... cy-rating/

Given these values, with a bit of math we can show that a player will break even on his two point field goal attempts if he hits on 30.4% of these shots. On three pointers the break-even point is 21.4%. If a player exceeds these thresholds, and virtually every NBA played does so with respect to two-point shots, the more he shoots the higher his value in PERs. So a player can be an inefficient scorer and simply inflate his value by taking a large number of shots.



If that would be true, we should find a very strong correlation between shots per minute and PER. Why shots per minute? Because PER is a per minute stats. The correlation coefficient between PER and shots per minute for all players from last season is an incredible 0.4. LOL, sorry, but Berri really has no real idea what he is talking about. He undervalues scoring, because he completetly forgot before he made his regression analysis that there is something like a 24 sec clock. If nobody takes a shot within those 24 sec, it will end up being a turnover. And someone has to take that shot to give the team at least a chance to score in this possession. His analysis for the last season for example had as an result that Erick Dampier is more valuable to the Mavericks than Dirk Nowitzki. Let alone that Dampier not even played as many minutes, because he couldn't stay on the floor (yeah, there is something like foul trouble too!), Dampier had a negative +/- value in comparison to Nowitzki's positive value. Berri really has a low understanding of the game itself, that is something you can really read out of his articles.
User avatar
BrooklynBulls
Retired Mod
Retired Mod
Posts: 32,733
And1: 2,652
Joined: May 13, 2007
Location: Avidly reading WillPenney.com
Contact:

Re: PER vs. Game Score vs. NBA Efficiency 

Post#33 » by BrooklynBulls » Sat Aug 15, 2009 11:26 am

PREACH BROTHA
hard49
Banned User
Posts: 939
And1: 1
Joined: Apr 28, 2002

Re: PER vs. Game Score vs. NBA Efficiency 

Post#34 » by hard49 » Sat Aug 15, 2009 1:27 pm

Berri's formula places to great an emphasis on rebounding while punishing players who have high amount of field goal attempts while discounting scoring.

It is a whacky formula...
User avatar
BallAboveAll
Bench Warmer
Posts: 1,307
And1: 4
Joined: Jun 02, 2007
Location: Los Angeles

Re: PER vs. Game Score vs. NBA Efficiency 

Post#35 » by BallAboveAll » Sat Aug 15, 2009 3:10 pm

Lmao @ the dumb response to more rings thing. Sure Horry isn't better than LBJ he was a role player on championship teams. You can't compare Horry to LeBron when it comes to winning. But you can compare Kobe and LeBron because they play similar roles as top dogs on contending teams.

I uderstand that rings don't make a player better then another. But stop with the dumb Horry>LBJ rebuttal statement.

As far as per and all that other stuff. I honestly don't keep track of it. Its cool that they came up with some of these stats. But the way some of these are used to praise players on here is a bit overboard at times.
Image
LebronsCavs
Banned User
Posts: 4,732
And1: 0
Joined: May 29, 2009

Re: PER vs. Game Score vs. NBA Efficiency 

Post#36 » by LebronsCavs » Sun Aug 16, 2009 1:04 am

nastneemz81 wrote:All it takes is one idiot to turn this into a Kobe/Lebron debate.

Just STFU they're both good. Better than everyone else who plays the game today.


I'm sorry but Wade is definitely a better player than Kobe nowadays. He is a more effective scorer, better play maker, better attacker of the basket, better defender. better teammate. He even shot at a similar clip from mid range this year (Kobe's bread and butter). Lastly, Wade shot a higher % from 3 pt range post ASG. So basically there really isnt any reason why anyone should think Kobe is a better individual player.
The top 2 players in the league are Lebron and Wade. Sorry kobe fans.
User avatar
amb1ent
Veteran
Posts: 2,810
And1: 17
Joined: May 22, 2006
Location: la ciudad
 

Re: PER vs. Game Score vs. NBA Efficiency 

Post#37 » by amb1ent » Sun Aug 16, 2009 1:14 am

i bet wade throws better sex parties than lebron and kobe combined, but i'm also positive that jordan's historical sex party record trumps anything wade could ever hope to accomplish.
Kayjay
Banned User
Posts: 4,839
And1: 180
Joined: Nov 30, 2008

Re: PER vs. Game Score vs. NBA Efficiency 

Post#38 » by Kayjay » Sun Aug 16, 2009 1:15 am

This really shoulda been Wins Produced vs. PER vs. Adjusted Plus-Minus.
User avatar
shobe_81
Inactive user
Inactive user
Posts: 2,749
And1: 0
Joined: Nov 17, 2007

Re: PER vs. Game Score vs. NBA Efficiency 

Post#39 » by shobe_81 » Sun Aug 16, 2009 1:21 am

Could Someone point to me what's Larry Bird's best PER?

1. Kobe will never be great at PER!
2. Lebroniez are afraid of Championships, most of them would have you believe that Lebron had the Greatest Playoffs performance EVER Because the PER says so. Oh What's that "Lebron didn't win the championship?
3. Using Horry to say Championships are pointless is falling into retardation when you're actually saying that a TOP 10 Players championships don't matter like as if they weren't a HUGE reason for the rings.

AGIAN What's Larry Bird's best PER?
Gutter92
Banned User
Posts: 1,864
And1: 2
Joined: May 28, 2008

Re: PER vs. Game Score vs. NBA Efficiency 

Post#40 » by Gutter92 » Sun Aug 16, 2009 1:23 am

twenty23 wrote:Kobe homers are scared of per.

:clap:

Return to The General Board