PER of greats only counting their prime

[Derivative]

I think the career PER is flawed because some players played a lot of seasons past their prime which reduced their average PER, so I did a PER average of the players top 5 seasons, which pretty much encompasses the players prime, which helps to measure how good a player is at their peak.

Michael Jordan: 31.08 PER
Shaquille O'Neal: 30.1
Kareem Abduljabbar 28.88
Wilt Chamberlain: 30.42
Bill Russell: 20.8
Larry Bird: 26.1
Magic Johnson: 26.26
Karl Malone: 27.46
Lebron James: 30.6
Kobe Bryant: 25.84
Tim Duncan: 26.82
Dwyane Wade: 28.6
Dirk Nowitzki: 26.4
Charles Barkley: 27.3
Hakeem Olajuwon: 25.86
Oscar Robertson: 26.34
Scottie Pippen: 21.92

[bwink23]

I think the career PER is flawed because some players played a lot of seasons past their prime which reduced their average PER, so I did a PER average of the players top 5 seasons, which pretty much encompasses the players prime, which helps to measure how good a player is at their peak.

Michael Jordan: 31.08 PER
Shaquille O'Neal: 30.1
Kareem Abduljabbar 28.88
Wilt Chamberlain: 30.42
Bill Russell: 20.8
Larry Bird: 26.1
Magic Johnson: 26.26
Karl Malone: 27.46
Lebron James: 30.6
Kobe Bryant: 25.84
Tim Duncan: 26.82
Dwyane Wade: 28.6
Dirk Nowitzki: 26.4
Charles Barkley: 27.3
Hakeem Olajuwon: 25.86
Oscar Robertson: 26.34
Scottie Pippen: 21.92

A player's absolute peak is usually their best 5-year stretch, not their best 5 years overall..

[Derivative]

A player's absolute peak is usually their best 5-year stretch, not their best 5 years overall..

what if a player is injured during middle of their peak, does that mean i have to add that injury riddled season to the peak calculator?

[bwink23]

No, not if they missed alot of games or something...you got to use your better judgment.

I wouldnt use years if they were say, 3-4 years apart.

[iamgine]

I think the career PER is flawed because some players played a lot of seasons past their prime which reduced their average PER, so I did a PER average of the players top 5 seasons, which pretty much encompasses the players prime, which helps to measure how good a player is at their peak.

Michael Jordan: 31.08 PER
Shaquille O'Neal: 30.1
Kareem Abduljabbar 28.88
Wilt Chamberlain: 30.42
Bill Russell: 20.8
Larry Bird: 26.1
Magic Johnson: 26.26
Karl Malone: 27.46
Lebron James: 30.6
Kobe Bryant: 25.84
Tim Duncan: 26.82
Dwyane Wade: 28.6
Dirk Nowitzki: 26.4
Charles Barkley: 27.3
Hakeem Olajuwon: 25.86
Oscar Robertson: 26.34
Scottie Pippen: 21.92
PER is flawed in the way it value scorers A LOT, maybe too much. Approximate 5 year prime PER:

Adrian Dantley: 24.5
Alex English: 22.5
George Gervin: 24
Dominique Wilkins: 24
Vince Carter: 23
Kiki Vandeweghe: 21.5
Mark Agguire: 22
Gilbert Arenas: 21
Amare Stoudemire: 25
Terry Cummings: 21

[Derivative]

PER is flawed in the way it value scorers A LOT, maybe too much. Approximate 5 year prime PER:

Adrian Dantley: 24.5
Alex English: 22.5
George Gervin: 24
Dominique Wilkins: 24
Vince Carter: 23
Kiki Vandeweghe: 21.5
Mark Agguire: 22
Gilbert Arenas: 21
Amare Stoudemire: 25
Terry Cummings: 21

i agree the number of assists is not weighted as heavy as efficient scoring, that's why good PGs don't have very high PER

[D-Wade316]

i agree the number of assists is not weighted as heavy as efficient scoring, that's why good PGs don't have very high PER
Chris Paul had a PER of 30 in 08-09. :pimp:

[Derivative]

Chris Paul had a PER of 30 in 08-09. :pimp:

o nvm then i take my statement back

[La Frescobaldi]

I think the career PER is flawed because some players played a lot of seasons past their prime which reduced their average PER, so I did a PER average of the players top 5 seasons, which pretty much encompasses the players prime, which helps to measure how good a player is at their peak.

Michael Jordan: 31.08 PER
Shaquille O'Neal: 30.1
Kareem Abduljabbar 28.88
Wilt Chamberlain: 30.42
Bill Russell: 20.8
Larry Bird: 26.1
Magic Johnson: 26.26
Karl Malone: 27.46
Lebron James: 30.6
Kobe Bryant: 25.84
Tim Duncan: 26.82
Dwyane Wade: 28.6
Dirk Nowitzki: 26.4
Charles Barkley: 27.3
Hakeem Olajuwon: 25.86
Oscar Robertson: 26.34
Scottie Pippen: 21.92


Steals and blocks were not counted before 1974. Both of those statistics are heavily weighted in the PER model.

According to PER, Wilt Chamberlain had the same number of blocked shots and steals as Bob Hogsett. Ever hear of Bob? Neither has anybody else. He played in a total of 7 games for the Pistons.

But yet, using PER, both those guys have the exact same number of blocked shots as Bill Russell, and the same number of steals as K.C. Jones. And they all have the same number of steals and blocked shots as Oscar Robertson... and every other player before 1974.

Zero.

PER is a good stat for comparing players before 1974, and it's a good stat for measuring players AFTER 1974.

Nobody who knows basketball would ever use PER to compare players before 1974 with players after that season.


This is exactly why I consider 1974 to be the beginning of the 3rd era of the NBA:

1946-1960
1960 - 1974
1974 - 1988
1988 - 2007
2007-today

Very simply...
Early league struggles,
Era of the Big Man,
1974 stats change when most stars retired or faded... 1988 the refs quit calling fouls.
And 2007 the catastrophe of Tim Donaghy scandal re-introduced the rules of basketball.

[magnax1]

Chris Paul had a PER of 30 in 08-09. :pimp:
He's the exception to the rule. Nash only had a low mid 20's PER in his best years, and if IIRC Isiah and Stockton weren't much different.

[jrong]

He's the exception to the rule. Nash only had a low mid 20's PER in his best years, and if IIRC Isiah and Stockton weren't much different.

That's because Chris Paul is a historically efficient PG. His PER and other advanced stats are so high because his turnovers are insanely low, and he maintains extremely high shooting percentages.

[Derivative]

That's because Chris Paul is a historically efficient PG. His PER and other advanced stats are so high because his turnovers are insanely low, and he maintains extremely high shooting percentages.

keep in mind PER adjusts to pace, nash and magic had relatively low PER because they played in very high pace teams which inflated stats. but CP3 played in a low pace team and still achieved the same stats

[jrong]

keep in mind PER adjusts to pace, nash and magic had relatively low PER because they played in very high pace teams which inflated stats. but CP3 played in a low pace team and still achieved the same stats

Well, I won't disparage Magic because he's earned the right to be considered above reproach. But, you're right, PER is a window into how much the pace factor made/makes Nash look better than he is.

[iamgine]

i agree the number of assists is not weighted as heavy as efficient scoring, that's why good PGs don't have very high PER
That's not the point. PER accounts for offense a lot but leave defensive lacking. Thus, it's not a good tool to measure greatness. Offensive efficiency and scoring greatness maybe, but that's about it.

Hakeem, Bill and Pippen would be many points higher if their defense is accounted for while Magic and Larry would decrease.

[D-Wade316]

That's not the point. PER accounts for offense a lot but leave defensive lacking. Thus, it's not a good tool to measure greatness. Offensive efficiency and scoring greatness maybe, but that's about it.

Hakeem, Bill and Pippen would be many points higher if their defense is accounted for while Magic and Larry would decrease.
Yeah. PER is strictly based on box scores. Defense is another. Any intelligent observer knows PER is just another stat used for comparison.

[magnax1]

That's because Chris Paul is a historically efficient PG. His PER and other advanced stats are so high because his turnovers are insanely low, and he maintains extremely high shooting percentages.
The TO's thing is true, but Nash was more efficient as a scorer, and scored a very similar amount on a per minute basis as CP3 (which is what PER uses) while also averaging a good deal more assists. Really, the CP3 vs Nash PER is just a window into PERs Flaws. It places way to much value on things like steals and turnovers. It's worth pointing out that the year CP3 averaged per 36 16 points on 58% TS% with 10 assists 2 steals and turnovers, he had the same PER as nash averaging per 36 19-12 on 65% TS% (which was league leading) with 4 turnovers and 1 steal. That's just a ludicrous assessment.

[Derivative]

The TO's thing is true, but Nash was more efficient as a scorer, and scored a very similar amount on a per minute basis as CP3 (which is what PER uses) while also averaging a good deal more assists. Really, the CP3 vs Nash PER is just a window into PERs Flaws. It places way to much value on things like steals and turnovers. It's worth pointing out that the year CP3 averaged per 36 16 points on 58% TS% with 10 assists 2 steals and turnovers, he had the same PER as nash averaging per 36 19-12 on 65% TS% (which was league leading) with 4 turnovers and 1 steal. That's just a ludicrous assessment.

cp3 has a higher per than nash because nash plays in a high pace team that inflates stats

hornets in 08-09 was 28th in the league in pace and averaged 95 ppg
while the suns consistant rank 1st in pace and average around 110 ppg

[Derivative]

bump

[fpliii]

One note...PER doesn't exist pre-77-78 season. It depends on blocked shots, steals, offensive rebounds and turnovers, which weren't all tracked until then.

basketball-reference.com ignores them in calculations:


The calcuation of uPER obviously depends on these statistics, so here are my solutions for years when the data are missing:

Zero out three-point field goals, turnovers, blocked shots, and steals.
Set the league value of possession (VOP) equal to 1.
Set the defensive rebound percentage (DRB%) equal to 0.7.
Set player offensive rebounds (ORB) equal to 0.3 * TRB.

source: http://www.basketball-reference.com/about/per.html

[Joyner82reload]

Durant's should be interesting. He's at 29.7 over the past 2 years, and at 31 for this season. For his best 4 years, which include his season as a 21 year old, he's at 28.0 PER. I would assume his top 5 seasons will be right around 30.

[Young X]

Chris Paul had a PER of 30 in 08-09. :pimp:Point GAWD

[SHAQisGOAT]

PER cannot be compared across eras, yet people continue to do it. :facepalm :facepalm :facepalm

Stop please.

[Derivative]

I think the career PER is flawed because some players played a lot of seasons past their prime which reduced their average PER, so I did a PER average of the players top 5 seasons, which pretty much encompasses the players prime, which helps to measure how good a player is at their peak.

Michael Jordan: 31.08 PER
Shaquille O'Neal: 30.1
Kareem Abduljabbar 28.88
Wilt Chamberlain: 30.42
Bill Russell: 20.8
Larry Bird: 26.1
Magic Johnson: 26.26
Karl Malone: 27.46
Lebron James: 30.6
Kobe Bryant: 25.84
Tim Duncan: 26.82
Dwyane Wade: 28.6
Dirk Nowitzki: 26.4
Charles Barkley: 27.3
Hakeem Olajuwon: 25.86
Oscar Robertson: 26.34
Scottie Pippen: 21.92

bump

[Droid101]

Derivative looking to get banned hard. 10 thread resurrections in 10 minutes? You're gone bro.

[Derivative]

Derivative looking to get banned hard. 10 thread resurrections in 10 minutes? You're gone bro.

only 8 bumps actually

[GimmeThat]

I suspect the variance of total PER by teams is greater than "offensive rating by team"

and I'm lazy.

[oarabbus]

Steals and blocks were not counted before 1974. Both of those statistics are heavily weighted in the PER model.

Nobody who knows basketball would ever use PER to compare players before 1974 with players after that season.


This is exactly why I consider 1974 to be the beginning of the 3rd era of the NBA:

1946-1960
1960 - 1974
1974 - 1988
1988 - 2007
2007-today



One note...PER doesn't exist pre-77-78 season. It depends on blocked shots, steals, offensive rebounds and turnovers, which weren't all tracked until then.

basketball-reference.com ignores them in calculations:



source: http://www.basketball-reference.com/about/per.html

So are you guys saying PER works to compare players pre-1974, between '74 and 77', and after '77, in 3 distinct groups?

[Force]

Only doing 5 years hurts the players who were better even longer. If you made it an 8 year stretch, the results would shuffle.

[pauk]

LeBron:

31.7
31.6
31.1
30.7
29.3

= 30.9............... not 30.6


Ironically you did calculate correctly on any other players though.... even then it still doesnt matter because "30.6" still puts him behind only Michael Jordan in any of those cases anyways.....

Also your order/ranking seems to be correct, but then it gets all messed up down there? If its no order, then why not have MJ at the bottom next to Malone aswell? :P

[Dragic4Life]

Legit top 10 ranking right there.:bowdown:

[Derivative]

I think the career PER is flawed because some players played a lot of seasons past their prime which reduced their average PER, so I did a PER average of the players top 5 seasons, which pretty much encompasses the players prime, which helps to measure how good a player is at their peak.

Michael Jordan: 31.08 PER
Shaquille O'Neal: 30.1
Kareem Abduljabbar 28.88
Wilt Chamberlain: 30.42
Bill Russell: 20.8
Larry Bird: 26.1
Magic Johnson: 26.26
Karl Malone: 27.46
Lebron James: 30.6
Kobe Bryant: 25.84
Tim Duncan: 26.82
Dwyane Wade: 28.6
Dirk Nowitzki: 26.4
Charles Barkley: 27.3
Hakeem Olajuwon: 25.86
Oscar Robertson: 26.34
Scottie Pippen: 21.92

bump

[dubeta]

So LeBron is the 2nd greatest ever? Sure.