Why do people who use traditional stats in argument discredit PER and the like?

[OldSkoolball#52]

Is it strictly about agenda? It seems that fans of chuckers are the ones who hate PER, win shares, TS% etc. the most, yet they have no problem flaunting 'raw' stats if that's what makes their hero look favorable.

Advanced stats just adjusts raw stats to move them closer to indicating a guy's impact.

I'd agree with anyone who says stats never tell the complete story on a guy. But there's really no denying that advanced stats at least give a better picture than traditional stats, and yet some of these agenda whores wholly dismiss any advanced metric, before proceeding to bleet "26 ppg!!!!" or whatever.


But agenda whores aside, and we know there are many, is there ANY reason to defer to traditional stats over advanced ones?

[branslowski]

A real NBA Fan who watches games: "He played great, 36pts 7reb 7ast 2stls 2blk 47%fg, especially in that 3rd quarter when the opposing team had the momentum, he hit 3 straight shots that pretty much stopped the others team run, so his team will enter the 4th up by 12 instead of up by 4". (This players team hits a 3 to start the 4th, teams up 15, and essentially opens up the game from there and wins in blowout fashion).....This fan watched the game, understood the context of the outcome, and appreciated the Original stats of the NBA (Pts, Reb, Ast)

How stat nerds cover games:

They don't watch it....A week goes by, they see a players stats: "wow, this guy had 25pts 7reb 12ast 57%fg shot the ball 12 times, PER at 32, winshare swag, him>>>the other guy"....Except the Real NBA fan watched that game, and saw Mr Winshares swag, stop shooting to protect a fg%, and then picked a quarter to only pass and get assist which occured in mostly garbage time due to his team building a lead while he was on the bench.


Fact/

[SamuraiSWISH]

Context:

Watch the games, don't read numbers and act like you know what you're talking about ...

[Owl]

A real NBA Fan who watches games: "He played great, 36pts 7reb 7ast 2stls 2blk 47%fg, especially in that 3rd quarter when the opposing team had the momentum, he hit 3 straight shots that pretty much stopped the others team run, so his team will enter the 4th up by 12 instead of up by 4". (This players team hits a 3 to start the 4th, teams up 15, and essentially opens up the game from there and wins in blowout fashion).....This fan watched the game, understood the context of the outcome, and appreciated the Original stats of the NBA (Pts, Reb, Ast)

How stat nerds cover games:

They don't watch it....A week goes by, they see a players stats: "wow, this guy had 25pts 7reb 12ast 57%fg shot the ball 12 times, PER at 32, winshare swag, him>>>the other guy"....Except the Real NBA fan watched that game, and saw Mr Winshares swag, stop shooting to protect a fg%, and then picked a quarter to only pass and get assist which occured in mostly garbage time due to his team building a lead while he was on the bench.


Fact/
Ignoring the bizzare assumption that watching games and believing in stats are mutually excluseive, the fact that people don't calculate or use all-in-one boxscore advanced stats for single games, the reality that all or almost all players don't not shoot to protect %s and the notion that points matter more for when they're scored ...

The NBA, in neither it's first year of existance as the BAA, nor its first year called the NBA did the league record rebounds. If you want "the original stats of the NBA" then you may utilise pts, assists, fga, fgm, fg%, fta, ftm, ft%, assists and fouls.

[OldSkoolball#52]

Still wondering.

[PJR]

The only contingent of fans who are THAT critical or completely disregard PER are Kobe fanatics. Because it accurately assesses their God. :oldlol:

[ZMonkey11]

They don't watch it....A week goes by, they see a players stats: "wow, this guy had 25pts 7reb 12ast 57%fg shot the ball 12 times, PER at 32, winshare swag, him>>>the other guy"....Except the Real NBA fan watched that game, and saw Mr Winshares swag, stop shooting to protect a fg%, and then picked a quarter to only pass and get assist which occured in mostly garbage time due to his team building a lead while he was on the bench.


Fact/

Although I understand your sentiment here, using assists to get your teammates involved, even in garbage time, builds team camaraderie. This improves morale for next game. This makes everyone play harder next game, harder picks for said player to score more on better FG% because they know when opportunity strikes, they will get the ball to pad their own stats to make that next big paycheck.

You are agenda building in a post that keys on agenda building. Get off it.

[Kblaze8855]

Advancted stats are pointless to me because they just combine numbers we already have.

If I have all the numbers that go into making PER...why the **** do I need some guy to decide for me what each is worth and combine them into one number? Who is anyone to determine the worth of an offensive rebound relative to say....a block? Really...what the **** is this:



uPER = (1 / MP) *
[ 3P
+ (2/3) * AST
+ (2 - factor * (team_AST / team_FG)) * FG
+ (FT *0.5 * (1 + (1 - (team_AST / team_FG)) + (2/3) * (team_AST / team_FG)))
- VOP * TOV
- VOP * DRB% * (FGA - FG)
- VOP * 0.44 * (0.44 + (0.56 * DRB%)) * (FTA - FT)
+ VOP * (1 - DRB%) * (TRB - ORB)
+ VOP * DRB% * ORB
+ VOP * STL
+ VOP * DRB% * BLK
- PF * ((lg_FT / lg_PF) - 0.44 * (lg_FTA / lg_PF) * VOP) ]


That has nothing to do with basketball....

[GOBB]

Advancted stats are pointless to me because they just combine numbers we already have.

If I have all the numbers that go into making PER...why the **** do I need some guy to decide for me what each is worth and combine them into one number? Who is anyone to determine the worth of an offensive rebound relative to say....a block? Really...what the **** is this:





That has nothing to do with basketball....


:roll: :roll: :roll:

[OldSkoolball#52]

Who is anyone to determine the worth of an offensive rebound relative to say....a block?


Well because there are computer programs these days that let you basically crunch every game in history and look for patterns. If an offensive rebounding advantage comes out as a much stronger correlation to winning than an advantage in blocks? Well....


I mean come on. Of course advanced stats aren't perfect. But they're better than traditional stats. They paint a more accurate picture. Of course the easiest way to make a judgment, if you actually know basketball, is to watch the games. But realistically what % of NBA games can a person watch in a season? There are 1,230 games played during the season. We all rely on stats to HELP make judgements, to some degree. Some do it more than others, but if you're going to do it.... why not just use the more accurate stats?

:confusedshrug:

[Heavincent]

The only contingent of fans who are THAT critical or completely disregard PER are Kobe fanatics. Because it accurately assesses their God. :oldlol:

The vast majority of of people disregard PER or don't even know what it is. It's entirely meaningless to everyone except for ESPN stat nerds and a few forum guys.

[Heavincent]

Advancted stats are pointless to me because they just combine numbers we already have.

If I have all the numbers that go into making PER...why the **** do I need some guy to decide for me what each is worth and combine them into one number? Who is anyone to determine the worth of an offensive rebound relative to say....a block? Really...what the **** is this:





That has nothing to do with basketball....

Yeah why the hell do people care about some stupid formula that John Hoolinger pulled out of his ass?

It's way too subjective to even be considered a statistic.

[Deuce Bigalow]

PER doesn't make a whole lot of sense really

Pau Gasol - 2010 Playoffs PER of 24.0
Larry Bird - 1986 Playoffs PER of 23.9
Magic Johnson - 1988 Playoffs PER of 22.9
Rick Barry - 1975 Playoffs PER of 22.5
Magic Johnson - 1982 Playoffs PER of 22.5
Magic Johbson - 1985 Playoffs PER of 22.3
Magic Johnson - 1980 Playoffs PER of 22.1
Pau Gasol - 2009 Playoffs PER of 21.9
Larry Bird - 1981 Playoffs PER of 21.8

I don't see how Pau was better than those guys.

Top 10 NBA PER

1. Michael Jordan* 27.91
2. LeBron James 27.70
3. Shaquille O'Neal 26.43
4. David Robinson* 26.18
5. Wilt Chamberlain* 26.13
6. Chris Paul 25.61
7. Dwyane Wade 25.46
8. Bob Pettit* 25.35
9. Neil Johnston* 24.69
10. Tim Duncan 24.65
11. Charles Barkley* 24.63
12. Kareem Abdul-Jabbar* 24.58
13. Magic Johnson* 24.11
14. Karl Malone* 23.90
15. Kevin Durant 23.67
16. Hakeem Olajuwon* 23.59
17. Larry Bird* 23.50
18. Dirk Nowitzki 23.46
19. Kobe Bryant 23.41
20. Oscar Robertson* 23.17

Makes an awful top 10. Kareem, Magic, Kobe, Bird, and Hakeem out of the top 10 and instead Chris Paul, Neil Johnston, Bob Pettit, and David Robinson are.

I can find the same problem for Win-shares.

[gts]

It's ok if you use it to compare a player against himself season to season or players on the same team playing in the same system but once you wander away from that criteria and start trying to use it to compare players from different eras or players that play in different systems you
are using the stat wrong and it becomes worthless noise

[OldSkoolball#52]

Top 10 NBA PER

1. Michael Jordan* 27.91
2. LeBron James 27.70
3. Shaquille O'Neal 26.43
4. David Robinson* 26.18
5. Wilt Chamberlain* 26.13
6. Chris Paul 25.61
7. Dwyane Wade 25.46
8. Bob Pettit* 25.35
9. Neil Johnston* 24.69
10. Tim Duncan 24.65
11. Charles Barkley* 24.63
12. Kareem Abdul-Jabbar* 24.58
13. Magic Johnson* 24.11
14. Karl Malone* 23.90
15. Kevin Durant 23.67
16. Hakeem Olajuwon* 23.59
17. Larry Bird* 23.50
18. Dirk Nowitzki 23.46
19. Kobe Bryant 23.41
20. Oscar Robertson* 23.17

Makes an awful top 10. Kareem, Magic, Kobe, Bird, and Hakeem out of the top 10 and instead Chris Paul, Neil Johnston, Bob Pettit, and David Robinson are.


So you think it's a given that the most famous players are automatically the most statistically productive?

Remember, PER simply condenses a player's statistical impact into a measure of efficiency based on things like minutes, pace, league average etc.

It does not RANK players. It simply compares them more accurately statistically.


The guy who created PER would not claim that the "Top 10 PER = Top 10 NBA". You are trying to make that assertion on your own, and it is not what PER was created for.



But I would also argue that your conventional assumption that only the most famous players can make up the actual Top 10 in impact is wrong. The idea that David Robinson could be one of the top 10 players in the league is not that silly. I mean think about what actually influences your perception of the Top 10. I guarantee it's based mostly on other peoples opinions, and the number of championships a guy has. Have you really sat down and watched film of Bird, Hakeem, and Robinson and tried to compare all of them objectively? Of course not.

There's also not a single numerical metric that suggests Kobe > Barkley but because every idiot watches PTI and repeatd what Tony Kornheiser says (i.e. "Rings = player value") we are left to force Kobe above Barkley based on the prestige that comes with team achievements. It's a horribly inadequate analysis, but it is basically the "lowest common denominator" analysis. It's one that is simple enough for everyone to use and participate in a discussion with. So it becomes the norm. Doesn't make it meaningful.

[Kblaze8855]

Well because there are computer programs these days that let you basically crunch every game in history and look for patterns. If a rebounding advantage comes out as a much stronger correlation to winning than an advantage in blocks? Well....


I mean come on. Of course advanced stats aren't perfect. But they're better than traditional stats. They paint a more accurate picture. Of course the easiest way to make a judgment, if you actually know basketball, is to watch the games. But realistically what % of NBA games can a person watch in a season? There are 1,230 games played during the season. We all rely on stats to HELP make judgements, to some degree. Some do it more than others, but if you're going to do it.... why not just use the more accurate stats?

Not all rebounds are created equally and half the blocks people get are because they failed to deny a drive/position in the first place....

Often times players shoot worse because the yare more talented than others and thus leaned on to make things happen....

Steals and blocks arent defense...

You cant account for ending a run, a shot to put a team up as opposed to one when the yare already up 20, or passes that arent assists.

Outlet passes, boxing out, getting back on defense, setting screens, the difference between being able to score 30 because your team put you in position for many easy baskets that you cant get on command and being a player who can make a shot without a defensive breakdown, and on and on and on.

You cant count it.

All we have are a few basic numbers...almost all of them misleading to begin with if you assume they directly correlate with the skill they are assigned to measure....

And people think its a good idea to combine 10-15 flawed numbers...while some guy decides what fraction a defensive rebound is worth and then try to apply pace which adds a whole other element that cant really be judged...because pace doesnt set itself.

We punsi hup tempo players...including point guards who are ASKED to make it uptempo. Sure they have more possessions...because they are told to make it a track meet. That may make it more likely they stack numbers...it doesnt mean their numbers need to be adjusted. They are a reflection of the style the coaching staff asked them to play. If it wins games....what sense does it make to edit down anything they did? The obvious answer...is to not assume the numbers they put up decide how good they are in the first place.


So many players get ****ed over by these numbers.


Isiah Thomas runs an offense that is just blowing the league away....hes having strings of games...10 plus games...where he puts up like 22 and 18. Hes getting 14 assists a night. Later hes asked to slow down and run an offense that can get good shots by grinding it out...isolations....feeding the post. So his numbers drop.

His team wins 2 rings and almost 3.

But....since he was now in a running league with guys who still played the way he used to....while he was slowing it down...his numbers arent as impressive. Sure its nice to be 18/9. But with Stockton doing 17/15, Timmy 23/11, KJ 22/12, Magic 24/12, Michael adams 27/11...its...not gonna stand out.

He could well have kept doing 22/14 through the entire 80s. But he takes a step back, runs his team, sets the tone, posts up Edwards, Aguirre, and so on....goes entire quarters without attacking. Dude plays MASTERFUL basketball.

So guuuuuess what? PER of 17.1.

Less than Emeka Okafor in his prime. Or for those who says you cant compare it in different seasons....about the same as Dell Curry putting up an efficient 12 a game on a 20 win Hornets team.

Isiah Thomas is the heart and soul....big shot maker...toughest player desipte being 6'1'', the playmaker, play caller, tempo setter, and unquestioned leader when shit gets real...of a two time NBA champion.

But because he took a gang of bailout shots and let his numbers slip for the good of the team hes Dell ****ing Curry to advanced stats.

I do not need that shit.

You want advanced stats...im not opposed. Lets get NEW numbers. Not numbers rearranged according to some.....guy.

Give me "advanced stats" to track fouls drawn in the bonus, screens set before a score, outlet passing leading to transition baskets, how many drives a player DIDNT need to try to block because he rotated in time.....give me a number that can account for shots missed when passed the ball with under 5 seconds on the clock....

Figure out how to put a number on ball denial.

Those are advanced stats.

PER?

Per isnt advanced.

PER is a cluster**** of numbers that are misleading on their own and combine to be even worse.

Drew ****ing Gooden over Ben Wallace when Ben is leading his team to the finals because defense cant be measured, Ben plays too many minutes, and Drew can make freethrows and be left open with Lebron and Z drawing attention away from where he likes to operate....

If only because of defense...PER can kiss my ass.

Combining a bunch of ****ed up numbers doesnt smooth out the errors in each individual one. It just makes for one massive bumpy ball of bullshit.

[Yao Ming's Foot]

Raw stats don't matter but if you manipulate them with an arbitrary formula they are infailable...

:facepalm

[Deuce Bigalow]

It's ok if you use it to compare a player against himself season to season or players on the same team playing in the same system but once you wander away from that criteria and start trying to use it to compare players from different eras or players that play in different systems you
are using the stat wrong and it becomes worthless noise
PER accounts for pace though.

Let's look at players on the same team during the same year.

2010 Playoffs
Kobe - 24.7 PER
Pau - 24.0 PER

They were basically equal? Okay...

[GOBB]

I guess starface isnt wondering anymore. Dinner has been served...


http://troll.me/images/jesse-pinkman/jesse-pinkman.jpg-"Bitch!"

[AirFederer]

So you think it's a given that the most famous players are automatically the most statistically productive?

Remember, PER simply condenses a player's statistical impact into a measure of efficiency based on things like minutes, pace, league average etc.

It does not RANK players. It simply compares them more accurately statistically.


The guy who created PER would not claim that the "Top 10 PER = Top 10 NBA". You are trying to make that assertion on your own, and it is not what PER was created for.



But I would also argue that your conventional assumption that only the most famous players can make up the actual Top 10 in impact is wrong. The idea that David Robinson could be one of the top 10 players in the league is not that silly. I mean think about what actually influences your perception of the Top 10. I guarantee it's based mostly on other peoples opinions, and the number of championships a guy has. Have you really sat down and watched film of Bird, Hakeem, and Robinson and tried to compare all of them objectively? Of course not.

There's also not a single numerical metric that suggests Kobe > Barkley but because every idiot watches PTI and repeatd what Tony Kornheiser says (i.e. "Rings = player value") we are left to force Kobe above Barkley based on the prestige that comes with team achievements. It's a horribly inadequate analysis, but it is basically the "lowest common denominator" analysis. It's one that is simple enough for everyone to use and participate in a discussion with. So it becomes the norm. Doesn't make it meaningful.

The Truth :applause:

[OldSkoolball#52]

Not all rebounds are created equally and half the blocks people get are because they failed to deny a drive/position in the first place....

Often times players shoot worse because the yare more talented than others and thus leaned on to make things happen....

Steals and blocks arent defense...

You cant account for ending a run, a shot to put a team up as opposed to one when the yare already up 20, or passes that arent assists.

Outlet passes, boxing out, getting back on defense, setting screens, the difference between being able to score 30 because your team put you in position for many easy baskets that you cant get on command and being a player who can make a shot without a defensive breakdown, and on and on and on.

You cant count it.

All we have are a few basic numbers...almost all of them misleading to begin with if you assume they directly correlate with the skill they are assigned to measure....

And people think its a good idea to combine 10-15 flawed numbers...while some guy decides what fraction a defensive rebound is worth and then try to apply pace which adds a whole other element that cant really be judged...because pace doesnt set itself.

We punsi hup tempo players...including point guards who are ASKED to make it uptempo. Sure they have more possessions...because they are told to make it a track meet. That may make it more likely they stack numbers...it doesnt mean their numbers need to be adjusted. They are a reflection of the style the coaching staff asked them to play. If it wins games....what sense does it make to edit down anything they did? The obvious answer...is to not assume the numbers they put up decide how good they are in the first place.


So many players get ****ed over by these numbers.


Isiah Thomas runs an offense that is just blowing the league away....hes having strings of games...10 plus games...where he puts up like 22 and 18. Hes getting 14 assists a night. Later hes asked to slow down and run an offense that can get good shots by grinding it out...isolations....feeding the post. So his numbers drop.

His team wins 2 rings and almost 3.

But....since he was now in a running league with guys who still played the way he used to....while he was slowing it down...his numbers arent as impressive. Sure its nice to be 18/9. But with Stockton doing 17/15, Timmy 23/11, KJ 22/12, Magic 24/12, Michael adams 27/11...its...not gonna stand out.

He could well have kept doing 22/14 through the entire 80s. But he takes a step back, runs his team, sets the tone, posts up Edwards, Aguirre, and so on....goes entire quarters without attacking. Dude plays MASTERFUL basketball.

So guuuuuess what? PER of 17.1.

Less than Emeka Okafor in his prime. Or for those who says you cant compare it in different seasons....about the same as Dell Curry putting up an efficient 12 a game on a 20 win Hornets team.

Isiah Thomas is the heart and soul....big shot maker...toughest player desipte being 6'1'', the playmaker, play caller, tempo setter, and unquestioned leader when shit gets real...of a two time NBA champion.

But because he took a gang of bailout shots and let his numbers slip for the good of the team hes Dell ****ing Curry to advanced stats.

I do not need that shit.

You want advanced stats...im not opposed. Lets get NEW numbers. Not numbers rearranged according to some.....guy.

Give me "advanced stats" to track fouls drawn in the bonus, screens set before a score, outlet passing leading to transition baskets, how many drives a player DIDNT need to try to block because he rotated in time.....give me a number that can account for shots missed when passed the ball with under 5 seconds on the clock....

Figure out how to put a number on ball denial.

Those are advanced stats.

PER?

Per isnt advanced.

PER is a cluster**** of numbers that are misleading on their own and combine to be even worse.

Drew ****ing Gooden over Ben Wallace when Ben is leading his team to the finals because defense cant be measured, Ben plays too many minutes, and Drew can make freethrows and be left open with Lebron and Z drawing attention away from where he likes to operate....

If only because of defense...PER can kiss my ass.

Combining a bunch of ****ed up numbers doesnt smooth out the errors in each individual one. It just makes for one massive bumpy ball of bullshit.


This is all well and good. But none of it actually disagrees with anything I've personally said in this thread.

My question to you is do you ever use traditional stats when talking about a player, or comparing players? If you have, or if you do.... then it kind of renders your opinion of PER rather hypocritical.


The people I'm questioning are those who will prop up a player based on selective traditional stats, then when it is pointed out that there PER does not translate as well, they're like "Pfffft, PER is stupid, means nothing, who cares about stat nerds"

[Kblaze8855]

If I like salt and pepper on my food....but dont like salt, pepper, and the 35 other things that make up shitty dollar store seasoning salt...

It doesnt make me a hypocrite.

PER containing an adjusted version of standard numbers doesnt mean you use both or neither.

"Raw" stats are what actually happened(barring manipulation which we all know happens). Per is someones idea of what each thing that happened is worth.

They are not the same.

[Rose'sACL]

If I like salt and pepper on my food....but dont like salt, pepper, and the 35 other things that make up shitty dollar store seasoning salt...

It doesnt make me a hypocrite.

PER containing an adjusted version of standard numbers doesnt mean you use both or neither.

"Raw" stats are what actually happened(barring manipulation which we all know happens). Per is someones idea of what each thing that happened is worth.

They are not the same.
While I don't like per that much, ts% is clearly a better stat than fg%.

[Deuce Bigalow]

So you think it's a given that the most famous players are automatically the most statistically productive?

Remember, PER simply condenses a player's statistical impact into a measure of efficiency based on things like minutes, pace, league average etc.

It does not RANK players. It simply compares them more accurately statistically.


The guy who created PER would not claim that the "Top 10 PER = Top 10 NBA". You are trying to make that assertion on your own, and it is not what PER was created for.



But I would also argue that your conventional assumption that only the most famous players can make up the actual Top 10 in impact is wrong. The idea that David Robinson could be one of the top 10 players in the league is not that silly. I mean think about what actually influences your perception of the Top 10. I guarantee it's based mostly on other peoples opinions, and the number of championships a guy has. Have you really sat down and watched film of Bird, Hakeem, and Robinson and tried to compare all of them objectively? Of course not.

There's also not a single numerical metric that suggests Kobe > Barkley but because every idiot watches PTI and repeatd what Tony Kornheiser says (i.e. "Rings = player value") we are left to force Kobe above Barkley based on the prestige that comes with team achievements. It's a horribly inadequate analysis, but it is basically the "lowest common denominator" analysis. It's one that is simple enough for everyone to use and participate in a discussion with. So it becomes the norm. Doesn't make it meaningful.
Championships may not matter to you but it matters to players and to me. Championships only matter when players are of similar caliber and have similar contributions to those championships. So of course team accomplishments don't matter when comparing someone like Steve Kerr to Charles Barkley, but when comparing Kobe to Barkley, how do you not factor in the championships? Okay, Barkley may not have had similar chances as Kobe had to win rings, but you can't penalize Kobe for actually winning. You can easily rank Kobe ahead of Barkley solely on the fact that Kobe was the more complete player on both sides of the court, without even looking at rings.

[gts]

PER accounts for pace though.

Let's look at players on the same team during the same year.

2010 Playoffs
Kobe - 24.7 PER
Pau - 24.0 PER

They were basically equal? Okay... who mentioned pace? I said systems... bball systems are not identical except one team plays faster than the other, there's a lot more than just pace to differentiate bball systems....

once again you have to use in in context, they play different roles within the team, why would you compare a center to a shooting guard, their shots are different their defensive assignments are different... Pau should actually be higher than Kobe considering his role...

[OldSkoolball#52]

So of course team accomplishments don't matter when comparing someone like Steve Kerr to Charles Barkley, but when comparing Kobe to Barkley, how do you not factor in the championships? Okay, Barkley may not have had similar chances as Kobe had to win rings, but you can't penalize Kobe for actually winning.

But "Kobe" didn't win. "Jordan" didn't win, "Magic" didn't win, "Duncan" didn't win.

The Lakers won. The Bulls won. The Spurs won.

Every organization has had superstars. The difference that separates franchises is primarily their ability to accumulate talent together, and also their willingness to spend what it takes to assemble and maintain that talent.

Winning is the result of a collective effort. No PLAYER wins basketball games. Barkley played for a cheap team. Is that his fault? If he'd have gone to a different team to play with Shaq, people would call him a ring chaser like they do with Lebron. The fact that Kobe was LUCKY enough to play for the Lakers who signed Shaquille O'Neal does not make him a better basketball player than Barkley. It has absolutely nothing to do with with it. It has literally zero weight in a logical comparison. If you think he was better based on game, fine. State your case when the issue comes up. But using "rings" is a complete cop out. It's the falsest of false arguments. It's designed for its simplicity, so that the average rube can use it and put in his 2 cents. Actual analysis excludes too many people, so "rings" is something for the simpletons. Those who actually prefer intelligent discussion laugh at the mere idea of it.

Period.

[Kblaze8855]

While I don't like per that much, ts% is clearly a better stat than fg%.

Same basic situation as per. If I have the numbers that combine to make it....I dont need anyone to combine them.

If I want to know why someones TS% is high I still have to look up exactly what they are doing far as overall shooting, from 3, and the Ft line. How much they are shooting and from where....

I can just take the numbers as they are.

I know Dwight cant shoot FTS, Kobe takes crazy fadeaways and doesnt shoot that well but makes his FTs, and that Deandre Jordan just dunks.

I dont need anything mixed up for me.

[gts]

While I don't like per that much, ts% is clearly a better stat than fg%.TS% is ok but it's too much work... you have to figure out why his TS% is what it is... you can't just say player xyz is a better shooter look at his TS% because in reality he could be a very mediocre shooter but really great from the stripe or all his shots are slam dunks but the guy hasn't hit a free throw in 5 years...lol too much work sorting out the why's to the numbers

[PJR]

But "Kobe" didn't win. "Jordan" didn't win, "Magic" didn't win, "Duncan" didn't win.

The Lakers won. The Bulls won. The Spurs won.

Every organization has had superstars. The difference that separates franchises is primarily their ability to accumulate talent together, and also their willingness to spend what it takes to assemble and maintain that talent.

Winning is the result of a collective effort. No PLAYER wins basketball games. Barkley played for a cheap team. Is that his fault? If he'd have gone to a different team to play with Shaq, people would call him a ring chaser like they do with Lebron. The fact that Kobe was LUCKY enough to play for the Lakers who signed Shaquille O'Neal does not make him a better basketball player than Barkley. It has absolutely nothing to do with with it. It has literally zero weight in a logical comparison. If you think he was better based on game, fine. State your case when the issue comes up. But using "rings" is a complete cop out. It's the falsest of false arguments. It's designed for its simplicity, so that the average rube can use it and put in his 2 cents. Actual analysis excludes too many people, so "rings" is something for the simpletons. Those who actually prefer intelligent discussion laugh at the mere idea of it.

Period.

Ether. :applause:

[Kblaze8855]

Exactly...if I still have to go do more work...why give me the formula number in the first place? It doesnt make things more simple. It gives one number with no information on how it came to be.

I guess its simple in that it is just...one number. But its not simple if you care about a players shooting and why the numbers are what the yare. And if you dont you wouldnt be posting them to begin with.

[Deuce Bigalow]

But "Kobe" didn't win. "Jordan" didn't win, "Magic" didn't win, "Duncan" didn't win.

The Lakers won. The Bulls won. The Spurs won.

Every organization has had superstars. The difference that separates franchises is primarily their ability to accumulate talent together, and also their willingness to spend what it takes to assemble and maintain that talent.

Winning is the result of a collective effort. No PLAYER wins basketball games. Barkley played for a cheap team. Is that his fault? If he'd have gone to a different team to play with Shaq, people would call him a ring chaser like they do with Lebron. The fact that Kobe was LUCKY enough to play for the Lakers who signed Shaquille O'Neal does not make him a better basketball player than Barkley. It has absolutely nothing to do with with it. It has literally zero weight in a logical comparison. If you think he was better based on game, fine. State your case when the issue comes up. But using "rings" is a complete cop out. It's the falsest of false arguments. It's designed for its simplicity, so that the average rube can use it and put in his 2 cents. Actual analysis excludes too many people, so "rings" is something for the simpletons. Those who actually prefer intelligent discussion laugh at the mere idea of it.

Period.
I agree. But you can't penalize Jordan, Kobe, Shaq, Duncan for winning. They had a good team so it doesn't matter that they won. Yes it does matter because their team accomplished the ultimate goal. Using any one thing - such as a single accomishments, a single stat to rank a player over another is not accurate. It's looking at everything, and championships are part of looking at everything. There are a lot of players that have rings but are not considered as great as ringless players. Look at players like Havlicek, Cowens, Pippen, Willis Reed, they are not considered better than guys like Baylor, West, and Barkley even though they all have 2 or more rings compared to zero for the others.

[moe94]

I agree. But you can't penalize Jordan, Kobe, Shaq, Duncan for winning. They had a good team so it doesn't matter that they won. Yes it does matter because their team accomplished the ultimate goal. Using any one thing - such as a single accomishments, a single stat to rank a player over another is not accurate. It's looking at everything, and championships are part of looking at everything.

Championships are strictly a team effort and have absolutely no bushiness in dictating an individual player's skill.

Horry somehow found himself as a part of, sometimes integral, the championships of three different organizations. That's not luck. Maybe he has this "winning" gene. :rolleyes:

[bdreason]

Because PER is a subjective advanced statistic, with different weights placed on statistics based on the opinion of one man.

[OldSkoolball#52]

I agree. But you can't penalize Jordan, Kobe, Shaq, Duncan for winning.

What do you mean penalize them? Let's say you and I are comparing Barkley and Kobe.

I would say "I prefer Barkley because he scored more efficiently while still pouring in points, and his above-average rebounding compared to other power forwards was an asset"

You might say "I prefer Kobe because I think he played better defense and could space the floor better from the three point line" or whatever you might say.

We are comparing them as players. Their impact on the game. I wouldn't say "Yeah, well Kobe was on the Lakers and that team won titles, so it takes away from this or that from Kobe".

That would be penalizing Kobe for the Lakers winning titles. Nobody is doing that. That is the converse of saying "Yeah, well Kobe was on the Lakers and that team won titles, so it takes away this or that from Barkley". Now you're penalizing Barkley because the Lakers won titles.





My point is that when you make comparisons, you have to minimize variables. That's why scientists do experiments. They don't just pick a reason and go with it. They break things down by process of elimination and find the root cause. Comparing players is subjective. You and I could have reasons for preferring different guys. But the reasons have to be relevant, they have to compare the players, not their organizations.

[moe94]

Is it true when PER was first formulated, it suggested Robinson and Hakeem had the GOAT peaks and thus it was remade to show how Jordan was really the best?

[OldSkoolball#52]

Because PER is a subjective advanced statistic, with different weights placed on statistics based on the opinion of one man.



Ok. I can accept this as a viewpoint.

However - statistics, PARTICULARLY points per game has been shown through the analysis of numerous statisticians to be a poor measure of a players impact.

Hollinger's analysis has been shown by the analysis of numerous statisticians to be a better measure of a players impact.

Hollinger didn't just drop acid and pull things out of the air. He looked through the statistical archives and found patterns and indicators that give a better description of what a player is doing than just raw stats. He gives an explanation for the formula and why each item is included. This has been "peer reviewed" by others in the field and found to be competent.


Again, I'm not saying PER is perfect by any stretch. Neither am I saying raw stats are perfect. But it's been demonstrated that PER is at least as accurate as raw stats when determining how much impact a player has offensively. So why the resistance to it by those who typically rely on raw stats to make their case?

[Deuce Bigalow]

What do you mean penalize them? Let's say you and I are comparing Barkley and Kobe.

I would say "I prefer Barkley because he scored more efficiently while still pouring in points, and his above-average rebounding compared to other power forwards was an asset"

You might say "I prefer Kobe because I think he played better defense and could space the floor better from the three point line" or whatever you might say.

We are comparing them as players. Their impact on the game. I wouldn't say "Yeah, well Kobe was on the Lakers and that team won titles, so it takes away from this or that from Kobe".

That would be penalizing Kobe for the Lakers winning titles. Nobody is doing that. That is the converse of saying "Yeah, well Kobe was on the Lakers and that team won titles, so it takes away this or that from Barkley". Now you're penalizing Barkley because the Lakers won titles.





My point is that when you make comparisons, you have to minimize variables. That's why scientists do experiments. They don't just pick a reason and go with it. They break things down by process of elimination and find the root cause. Comparing players is subjective. You and I could have reasons for preferring different guys. But for the reasons to be relevant, they have to compare the players, not their organizations.
I know Kobe can lead a team to win a championship when given a good supporting cast. I'm sure Barkley can too, but he isn't proven like Kobe. I know Kobe can take over games in the 4th quarter, as I have seen him do many times so spare me his clutch stats. I know Kobe can lock down a player in spurts. I also know that he will push his teammates in practice and that he makes his teammates better, no coincidence that many players had their best seasons with the Lakers, Phil Jackson was a factor too I'm sure. That is why I am safe taking him over Barkley, he has those things over him plus the other factors you mentioned.

[ballinhun8]

PER is a made up numeric number by a writer for ESPN.



If someone created his own formula for fox sports 1 and presented it where it did not take something like FG% as a vital stat, would we take that seriously??

[moe94]

I can understand not liking PER because you disagree with the formula, but why is the fact that some dude from ESPN made it the first thing that gets attacked? How is that relevant at all?

[Pursuer]

Ok. I can accept this as a viewpoint.

However - statistics, PARTICULARLY points per game has been shown through the analysis of numerous statisticians to be a poor measure of a players impact.

Hollinger's analysis has been shown by the analysis of numerous statisticians to be a better measure of a players impact.



You're comparing PPG vs PER here? I don't think anyone's saying PPG is a better measure of a players impact vs PER. I think the point everybody's trying to make here, is that PER contains subjective inserted weights, based on calculations, made on assumptions. PER is way too flawed, in terms of it's weight distributing nature. Players impact analysis can be done better with the raw stats vs PER, because PER only shows the assumptive weighted end result of a statistical line, where as raw stats show the parts and thus give a nature of the combination, not the combination itself. It's way to simplistic to assume anything regarding a player's impact is to be measure by a single number. I'm not saying it could be possible in the future, but eventually it will still stay subjective, because basketball, I think, won't be completely figured out to the point, where every single action has a perfect end result. That's why combining the elements is really just dropping your opinion(even if it's calculated using correlations).

[qrich]

Because PER is a subjective advanced statistic, with different weights placed on statistics based on the opinion of one man.

Basically. Anyone can go out and spend time making a random formula to try to define the impact a specific player has had.

But advance or raw statistic, the eye test should be just as vital.

[moe94]

But advance or raw statistic, the eye test should be just as vital.

You can't argue advanced metrics are useless because of their subjective nature and then champion "eye test".

[atljonesbro]

You can't argue advanced metrics are useless because of their subjective nature and then champion "eye test".
Exactly what is the eye test lmao. Jeff Teague>Cp3 shout out to the eye test.

[moe94]

Exactly what is the eye test lmao. Jeff Teague>Cp3 shout out to the eye test.
:oldlol:

[qrich]

You can't argue advanced metrics are useless because of their subjective nature and then champion "eye test".

Did I say useless, or did I say that they should be "just as vital."


Exactly what is the eye test lmao. Jeff Teague>Cp3 shout out to the eye test.

:rolleyes:

Explain how so bud.

I mean, Paul is a better floor general, can turn it up a notch when required, doesn't make many mistakes, gets everyone involved. But hey, if you can't comprehend, then, it is fine by me.

[atljonesbro]

Did I say useless, or did I say that they should be "just as vital."



:rolleyes:

Explain how so bud.

I mean, Paul is a better floor general, can turn it up a notch when required, doesn't make many mistakes, gets everyone involved. But hey, if you can't comprehend, then, it is fine by me.
The eye test don't question me. ALL HAIL THE ALMIGHTY EYE TEST :bowdown: :bowdown:

[qrich]

The eye test don't question me. ALL HAIL THE ALMIGHTY EYE TEST :bowdown: :bowdown:

It continues.

No Child Left Behind really showing how well the program worked.

[moe94]

It continues.

No Child Left Behind really showing how well the program worked.

I think what he's saying is that people really do try to use that as an excuse to suggest certain players are worse than they really are or better than they really are, without any real analysis or reasoning outside "I like how his game looks, aesthetically".

[OldSkoolball#52]

It continues.

No Child Left Behind really showing how well the program worked.


To be fair, I have a feeling he was left behind on purpose

[qrich]

I think what he's saying is that people really do try to use that as an excuse to suggest certain players are worse than they really are or better than they really are, without any real analysis or reasoning outside "I like how his game looks, aesthetically".

So saying just as vital means that.

Got it.

The lack of comprehension on this board never ceases to amaze me. Yet, its still superior to RealGM and what BasketballBoards.net turned into. If only the old BBB would make a comeback.

[Owl]

Not all rebounds are created equally and half the blocks people get are because they failed to deny a drive/position in the first place....

Often times players shoot worse because the yare more talented than others and thus leaned on to make things happen....

Steals and blocks arent defense...

You cant account for ending a run, a shot to put a team up as opposed to one when the yare already up 20, or passes that arent assists.

Outlet passes, boxing out, getting back on defense, setting screens, the difference between being able to score 30 because your team put you in position for many easy baskets that you cant get on command and being a player who can make a shot without a defensive breakdown, and on and on and on.

You cant count it.

All we have are a few basic numbers...almost all of them misleading to begin with if you assume they directly correlate with the skill they are assigned to measure....

And people think its a good idea to combine 10-15 flawed numbers...while some guy decides what fraction a defensive rebound is worth and then try to apply pace which adds a whole other element that cant really be judged...because pace doesnt set itself.

We punsi hup tempo players...including point guards who are ASKED to make it uptempo. Sure they have more possessions...because they are told to make it a track meet. That may make it more likely they stack numbers...it doesnt mean their numbers need to be adjusted. They are a reflection of the style the coaching staff asked them to play. If it wins games....what sense does it make to edit down anything they did? The obvious answer...is to not assume the numbers they put up decide how good they are in the first place.


So many players get ****ed over by these numbers.


Isiah Thomas runs an offense that is just blowing the league away....hes having strings of games...10 plus games...where he puts up like 22 and 18. Hes getting 14 assists a night. Later hes asked to slow down and run an offense that can get good shots by grinding it out...isolations....feeding the post. So his numbers drop.

His team wins 2 rings and almost 3.

But....since he was now in a running league with guys who still played the way he used to....while he was slowing it down...his numbers arent as impressive. Sure its nice to be 18/9. But with Stockton doing 17/15, Timmy 23/11, KJ 22/12, Magic 24/12, Michael adams 27/11...its...not gonna stand out.

He could well have kept doing 22/14 through the entire 80s. But he takes a step back, runs his team, sets the tone, posts up Edwards, Aguirre, and so on....goes entire quarters without attacking. Dude plays MASTERFUL basketball.

So guuuuuess what? PER of 17.1.

Less than Emeka Okafor in his prime. Or for those who says you cant compare it in different seasons....about the same as Dell Curry putting up an efficient 12 a game on a 20 win Hornets team.

Isiah Thomas is the heart and soul....big shot maker...toughest player desipte being 6'1'', the playmaker, play caller, tempo setter, and unquestioned leader when shit gets real...of a two time NBA champion.

But because he took a gang of bailout shots and let his numbers slip for the good of the team hes Dell ****ing Curry to advanced stats.

I do not need that shit.

You want advanced stats...im not opposed. Lets get NEW numbers. Not numbers rearranged according to some.....guy.

Give me "advanced stats" to track fouls drawn in the bonus, screens set before a score, outlet passing leading to transition baskets, how many drives a player DIDNT need to try to block because he rotated in time.....give me a number that can account for shots missed when passed the ball with under 5 seconds on the clock....

Figure out how to put a number on ball denial.

Those are advanced stats.

PER?

Per isnt advanced.

PER is a cluster**** of numbers that are misleading on their own and combine to be even worse.

Drew ****ing Gooden over Ben Wallace when Ben is leading his team to the finals because defense cant be measured, Ben plays too many minutes, and Drew can make freethrows and be left open with Lebron and Z drawing attention away from where he likes to operate....

If only because of defense...PER can kiss my ass.

Combining a bunch of ****ed up numbers doesnt smooth out the errors in each individual one. It just makes for one massive bumpy ball of bullshit.
No offense but there are contradictions in your argument.

OP says that they're as good as the raw stats that go into them. You say they aren't because raw stats are bad e.g. "not all rebounds are created equal" and "passes that aren't assists". So should we not count rebounds or assists too? Then there's Isiah gets way too punished for slowing down his team, yet "what sense does it make to edit down what they did". So Isiah is too punished for slowing down by this measure, yet what it should do is not account for pace at all which would make Isiah's drops look worse?

Steals and blocks aren't defense. PER's creator admits as much. On this basis, what you're against (or what you can justify being against) is the misuse of PER as an infallible tool to measure production absolutely. Incidentally Hollinger messed around with a defensive PER but I don't think he had play by play data and it was only looking at man, rather than team defense.

In terms of boxing out, picks, floor spacing etc. PER can't capture this. But
1) Again it's not meant to.
and
2) You can adjust based on your knowledge of the game.

Re Pace: It fundamentally comes down to, do you believe a basket is of the same value if it comes in 63-60 game than in a 126-120 game? Basketball is a game of possessions. In a given game teams have the same number of possessions. I don't care how many points my players score in absolute terms. And you give an example of why. Michael Adams played on crazy paced team. Should we take his and Orlando Woolridge's 25ppg season at face value or should we take account of pace. Not that PER does this best, but fundamentally should we account for context? Yes. And doing so systematically is most fair.

Isiah's numbers drop except his usage doesn't (so he was taking on the same proportion of the offense as before, a slightly higher one actually, than before the arrival of Edwards and Aguirre (and before him Dantley)) and his turnovers stay high and he gets to the line less. Isiah got worse, his teammates got better. He did improve in the playoffs, still the Pistons won with elite defense and Isiah wasn't a huge part of that. He was an important part of keeping the offense above average, especially in the playoffs.

Minutie based advanced stats are there. And they are useful. But big men clogging the lane and deterring drives is even more subjective than assists. And you may have to pay for that sort of analysis because it won't generate hits for big commerical sites so you need to go to niche places.

As before it's creator has repeatedly emphasized D isn't included so it's not a comprehensive, this is how good a player is stat. No one argued for Gooden as better than Wallace no matter how metrically inclined because they knew Wallace was an elite defender and Gooden broke plays. But most players don't differs as much as these two in non-boxscore aspects of the game so advanced stats do make useful ballpark figures and watching and using metrics aren't mutually exclusive.

Not that PER or any metric is perfect. PER overvalues scorers. I'm not sure any boxscore metric accounts well for the greater difficulty of scoring in the halfcourt. It's incomplete but then so are our memories unless you have a pefect photographic memory of every basketball game ever played and remained wholly uninfluenced by external media.


Same basic situation as per. If I have the numbers that combine to make it....I dont need anyone to combine them.

If I want to know why someones TS% is high I still have to look up exactly what they are doing far as overall shooting, from 3, and the Ft line. How much they are shooting and from where....

I can just take the numbers as they are.

I know Dwight cant shoot FTS, Kobe takes crazy fadeaways and doesnt shoot that well but makes his FTs, and that Deandre Jordan just dunks.

I dont need anything mixed up for me.
But you'd need to know not only the individual %s but the proportion of shots taken from each location, how often a player got to the line. It's like saying, I've got, eggs, flour, milk, sugar and an oven so why would I want that cake. Well it's already made for you, it's ready to eat straight away, it's free and you can still use your flour, milk etc if this cake hasn't met your cake based needs.

It's something that stops people saying that because Billups has a bad fg% (incidentally one of the rare instances where mixing together different factors of the game is for the worse, because it treats 2s and 3s as the same) he's an inefficient offensive player (he wasn't because he got to the line and took a lot of threes). Maybe you knew that anyway. That doesn't mean it's not useful to have something prove it, and it doesn't prevent you using it's constituent parts, it just means you don't have to do the math each time you want to combine them accurately.

[tpols]

Advanced stats that stay within a single pathway ain't bad.. Like points per possession for scoring.. Or assist percentage rebound percentage breaking down individual stats into relative numbers. But stats that lump everything together suck

[moe94]

So saying just as vital means that.

Got it.
.

No one is attacking you personally but the use of the "eye test" as an actual analysis. Relax. :rolleyes:

[kuniva_dAMiGhTy]

Because I don't need adjusted numbers to tell me what raw stats and watching games do. :confusedshrug:

"PER" is useful within context. I like that it adjusts for pace.

[Owl]

Basically. Anyone can go out and spend time making a random formula to try to define the impact a specific player has had.

But advance or raw statistic, the eye test should be just as vital.
But not anyone can get theirs published, or parlay it into a job as a team consultant or front office staff. Some numbers seem more credible because others have bought into them and backed that interest with cash.

And certainly watching games and using metrics should be complementary. Just as vital? It depends what for? Assessing a players ability/impact generally I'd take an advanced stat over one person's opinion/"eye test", though again, ideally you'd have both. If I was designing an offense you'd want to watch games and have minutie based stats (where is player x shooting well from). It's about having the right tool for the job.

[LAZERUSS]

No offense but there are contradictions in your argument.

OP says that they're as good as the raw stats that go into them. You say they aren't because raw stats are bad e.g. "not all rebounds are created equal" and "passes that aren't assists". So should we not count rebounds or assists too? Then there's Isiah gets way too punished for slowing down his team, yet "what sense does it make to edit down what they did". So Isiah is too punished for slowing down by this measure, yet what it should do is not account for pace at all which would make Isiah's drops look worse?

Steals and blocks aren't defense. PER's creator admits as much. On this basis, what you're against (or what you can justify being against) is the misuse of PER as an infallible tool to measure production absolutely. Incidentally Hollinger messed around with a defensive PER but I don't think he had play by play data and it was only looking at man, rather than team defense.

In terms of boxing out, picks, floor spacing etc. PER can't capture this. But
1) Again it's not meant to.
and
2) You can adjust based on your knowledge of the game.

Re Pace: It fundamentally comes down to, do you believe a basket is of the same value if it comes in 63-60 game than in a 126-120 game? Basketball is a game of possessions. In a given game teams have the same number of possessions. I don't care how many points my players score in absolute terms. And you give an example of why. Michael Adams played on crazy paced team. Should we take his and Orlando Woolridge's 25ppg season at face value or should we take account of pace. Not that PER does this best, but fundamentally should we account for context? Yes. And doing so systematically is most fair.

Isiah's numbers drop except his usage doesn't (so he was taking on the same proportion of the offense as before, a slightly higher one actually, than before the arrival of Edwards and Aguirre (and before him Dantley)) and his turnovers stay high and he gets to the line less. Isiah got worse, his teammates got better. He did improve in the playoffs, still the Pistons won with elite defense and Isiah wasn't a huge part of that. He was an important part of keeping the offense above average, especially in the playoffs.

Minutie based advanced stats are there. And they are useful. But big men clogging the lane and deterring drives is even more subjective than assists. And you may have to pay for that sort of analysis because it won't generate hits for big commerical sites so you need to go to niche places.

As before it's creator has repeatedly emphasized D isn't included so it's not a comprehensive, this is how good a player is stat. No one argued for Gooden as better than Wallace no matter how metrically inclined because they knew Wallace was an elite defender and Gooden broke plays. But most players don't differs as much as these two in non-boxscore aspects of the game so advanced stats do make useful ballpark figures and watching and using metrics aren't mutually exclusive.

Not that PER or any metric is perfect. PER overvalues scorers. I'm not sure any boxscore metric accounts well for the greater difficulty of scoring in the halfcourt. It's incomplete but then so are our memories unless you have a pefect photographic memory of every basketball game ever played and remained wholly uninfluenced by external media.


But you'd need to know not only the individual %s but the proportion of shots taken from each location, how often a player got to the line. It's like saying, I've got, eggs, flour, milk, sugar and an oven so why would I want that cake. Well it's already made for you, it's ready to eat straight away, it's free and you can still use your flour, milk etc if this cake hasn't met your cake based needs.

It's something that stops people saying that because Billups has a bad fg% (incidentally one of the rare instances where mixing together different factors of the game is for the worse, because it treats 2s and 3s as the same) he's an inefficient offensive player (he wasn't because he got to the line and took a lot of threes). Maybe you knew that anyway. That doesn't mean it's not useful to have something prove it, and it doesn't prevent you using it's constituent parts, it just means you don't have to do the math each time you want to combine them accurately.

Excellent post.

Personally, I believe that Advanced Stats, used with the actual stats, can justify a players overall contributions. They are certainly not an end-all, but just like the traditional stat lines, they are at least a good indicator of production.

However, some advanced stats are easily distorted. The PER/36 minutes stat basically punishes a player who plays 40-48 mpg, and rewards those that play less (Rodman's TRB% being a good indicator of that, or Howard's production.) Let's face reality, a player playing 45 mpg is generally going to be less efficient than a player playing 30.

And for those that use "pace"...please. It's bad enough that they are constantly used to disparage Wilt (even though no one else in his, or any other era, came close to his offensive domination.) But, I have also read those that diminish Oscar's triple-double seasons based on "pace." But, how does one disparage a player that averages over 30 ppg on 22 FGAs? Or hands out assists at way more than the league average, and obviously in an era when they were much tougher to come by. You can realistically reduce his rebounding numbers, but given all of the that, his numbers STILL translate into 30-8-10 seasons. And the "paceologists" seldom acknowledge league average's in their arguments. I have seen posters claim that Wilt was "inefficient" in his 50 ppg season, based on his .506 FG%. But they will never mention that the league average was .426. Or that he averaged 45 ppg on a .528 FG% the very next season, or 34 ppg on a .540 FG% the season after that, which blows away players like Hakeem and D-Rob in their greatest scoring seasons (and in league's that shot far greater.)

And of course, one only need to look at where Bill Russell ranks in all-time PER to see the obvious flaws with that statistic.

It really only takes some degree of common sense to understand a player's overall contributions. Players like KAJ and Worthy were huge beneficiary's of Magic's brilliance. As were Russell's Celtics, and Duncan's Spurs. And then players like Shaq, MJ, KAJ, and Wilt...all capable of single-handedly carrying teams. If anything, Chamberlain's numbers were even more staggering considering that he was the focal point of opposing team's defenses (as well as having horrible rosters early in his career, that contributed nothing.)

And finally, for those that completely ignore stats, you are among a tiny minority. Even the Vegas odds-makers use tons of stats, and their careers are dependent upon them. To simply say that a player has "inflated" stats (e.g Kevin Love's rebounding) is just being ridiculous. If there were really a true possibility of having inflated stats, virtually every team would have players putting up unbelievable numbers. It just doesn't happen over the course of an entire season.

[RoundMoundOfReb]

No offense but there are contradictions in your argument.

OP says that they're as good as the raw stats that go into them. You say they aren't because raw stats are bad e.g. "not all rebounds are created equal" and "passes that aren't assists". So should we not count rebounds or assists too? Then there's Isiah gets way too punished for slowing down his team, yet "what sense does it make to edit down what they did". So Isiah is too punished for slowing down by this measure, yet what it should do is not account for pace at all which would make Isiah's drops look worse?

Steals and blocks aren't defense. PER's creator admits as much. On this basis, what you're against (or what you can justify being against) is the misuse of PER as an infallible tool to measure production absolutely. Incidentally Hollinger messed around with a defensive PER but I don't think he had play by play data and it was only looking at man, rather than team defense.

In terms of boxing out, picks, floor spacing etc. PER can't capture this. But
1) Again it's not meant to.
and
2) You can adjust based on your knowledge of the game.

Re Pace: It fundamentally comes down to, do you believe a basket is of the same value if it comes in 63-60 game than in a 126-120 game? Basketball is a game of possessions. In a given game teams have the same number of possessions. I don't care how many points my players score in absolute terms. And you give an example of why. Michael Adams played on crazy paced team. Should we take his and Orlando Woolridge's 25ppg season at face value or should we take account of pace. Not that PER does this best, but fundamentally should we account for context? Yes. And doing so systematically is most fair.

Isiah's numbers drop except his usage doesn't (so he was taking on the same proportion of the offense as before, a slightly higher one actually, than before the arrival of Edwards and Aguirre (and before him Dantley)) and his turnovers stay high and he gets to the line less. Isiah got worse, his teammates got better. He did improve in the playoffs, still the Pistons won with elite defense and Isiah wasn't a huge part of that. He was an important part of keeping the offense above average, especially in the playoffs.

Minutie based advanced stats are there. And they are useful. But big men clogging the lane and deterring drives is even more subjective than assists. And you may have to pay for that sort of analysis because it won't generate hits for big commerical sites so you need to go to niche places.

As before it's creator has repeatedly emphasized D isn't included so it's not a comprehensive, this is how good a player is stat. No one argued for Gooden as better than Wallace no matter how metrically inclined because they knew Wallace was an elite defender and Gooden broke plays. But most players don't differs as much as these two in non-boxscore aspects of the game so advanced stats do make useful ballpark figures and watching and using metrics aren't mutually exclusive.

Not that PER or any metric is perfect. PER overvalues scorers. I'm not sure any boxscore metric accounts well for the greater difficulty of scoring in the halfcourt. It's incomplete but then so are our memories unless you have a pefect photographic memory of every basketball game ever played and remained wholly uninfluenced by external media.


But you'd need to know not only the individual %s but the proportion of shots taken from each location, how often a player got to the line. It's like saying, I've got, eggs, flour, milk, sugar and an oven so why would I want that cake. Well it's already made for you, it's ready to eat straight away, it's free and you can still use your flour, milk etc if this cake hasn't met your cake based needs.

It's something that stops people saying that because Billups has a bad fg% (incidentally one of the rare instances where mixing together different factors of the game is for the worse, because it treats 2s and 3s as the same) he's an inefficient offensive player (he wasn't because he got to the line and took a lot of threes). Maybe you knew that anyway. That doesn't mean it's not useful to have something prove it, and it doesn't prevent you using it's constituent parts, it just means you don't have to do the math each time you want to combine them accurately.


Basically summed up my thoughts well.


I'm pretty sure nobody says PER is the end all be all. Nobody says the player with the higher PER is always the better player. but it is certainly superior to raw stats like ppg,rpg,apg etc. Like how passer rating is superior to raw yardage.

[Kblaze8855]

A cake?

A cake only has to be good. Why it's good doesn't matter. ts%? it can be great and you are a terrible scorer. so why it's great matters.

A great cake is a success. Great ts..... can be despite you sucking.

Far as the other post. .... seems to be some fundamental misunderstandings. especially on Isiah. but I'm on a phone so. ... **** it

[OldSkoolball#52]

A great cake is a success. Great ts..... can be despite you sucking.



But if two guys have the same ppg and one of them has a much higher TS%... it's pretty telling.

That's what TS% is to be used for.

And it works effectively.

[RoundMoundOfReb]

It's all about using the stats in the proper context. Deandre Jordan has a higher ts% than Kobe but nobody would seriously say Deandre is a better scorer than Kobe cause all he does is make open, gift wrapped dunks where as Kobe creates his own offense and does it at a much higher volume.

[Young X]

The problem is people are looking at stats like PER as player ranking stats when they're PRODUCTION stats.

Player A having a higher PER than player B doesn't mean he's better, it means he's more productive per minute/possession. That's it. It doesn't rank players based on how many rings they have, how athletic they are, or their intangibles, but guess what? Regular stats don't either.

Kevin Love having a higher PER than someone like Bill Russell doesn't mean he's better, it means his production rate is higher, which is true. It's not that complicated.

PER = production rate. Understand what the stats mean before you dismiss them.

[tpols]

The problem is people are looking at stats like PER as player ranking stats when they're PRODUCTION stats.

Player A having a higher PER than player B doesn't mean he's better, it means he's more productive per minute/possession. That's it. It doesn't rank players based on how many rings they have, how athletic they are, or their intangibles, but guess what? Regular stats don't either.

Kevin Love having a higher PER than someone like Bill Russell doesn't mean he's better, it means his production rate is higher, which is true. It's not that complicated.

PER = production rate. Understand what the stats mean before you dismiss them.
Who cares if his production rate is higher when he has half the impact on a basketball game? :oldlol:

[OldSkoolball#52]

Who cares if his production rate is higher when he has half the impact on a basketball game? :oldlol:


Well here's the thing: You wouldn't want to use it for guys who are very dissimilar.

But you know how people say "Dude, Jordan and Horry's rings aren't the same. You can't use rings to compare one guy who's "the man" with another who's not "the man"!!!

It's pretty much like that. You're not gonna compare a defensive players PER to a scorer. But if you're comparing two primarily offensive players and want a snapshot of their effectiveness, PER is going to be as good or better than any simply discussion of points/assists/fg%


I mean how can anyone justify using "rings" to compare similar players, but not PER? Oh, right. Because they prefer the results when you use "rings" instead of PER.

It's no coincidence that Kobe stans are BY FAR the biggest detractors from advanced metrics. Because their hero, their favorite player, is by all analytical accounts the most overrated player in NBA history.

[Deuce Bigalow]

Well here's the thing: You wouldn't want to use it for guys who are very dissimilar.

But you know how people say "Dude, Jordan and Horry's rings aren't the same. You can't use rings to compare one guy who's "the man" with another who's not "the man"!!!

It's pretty much like that. You're not gonna compare a defensive players PER to a scorer. But if you're comparing two primarily offensive players and want a snapshot of their effectiveness, PER is going to be as good or better than any simply discussion of points/assists/fg%


I mean how can anyone justify using "rings" to compare similar players, but not PER? Oh, right. Because they prefer the results when you use "rings" instead of PER.

It's no coincidence that Kobe stans are BY FAR the biggest detractors from advanced metrics. Because their hero, their favorite player, is by all analytical accounts the most overrated player in NBA history.
Considering that Kobe has a higher PER than Larry Bird and Magic Johnson, no he isn't the most overrated.

[moe94]

Considering that Kobe has a higher PER than Larry Bird and Magic Johnson, no he isn't the most overrated.

The irony of this statement knows no bounds.

[Deuce Bigalow]

The irony of this statement knows no bounds.
I'm going by his logic.

[Micku]

Is it strictly about agenda? It seems that fans of chuckers are the ones who hate PER, win shares, TS% etc. the most, yet they have no problem flaunting 'raw' stats if that's what makes their hero look favorable.

...

But agenda whores aside, and we know there are many, is there ANY reason to defer to traditional stats over advanced ones?

It could be a mixture. Like you mentioned, stats don't tell the full story and you have to take it within context. Like TS% just combines the FTs, 3pt, and FG%. But it doesn't tell how the player does on the floor when the defense is active. Like a guy like Kevin Durant. His stat sheet a couple nights ago. His FG was: 3-9. His 3pt was 1-1. His FTs was 12-13. He got 19 points overall.

So, his FG% was 33%. His eFG% was 38.9%. His TS% was 64.5%. Why? Because he was 1-1 from 3 and he went to the FT and is a good FT shooter. Did he have a good night because of his high TS%? No. He couldn't score while the defense was live, but he got his points from the FT. It doesn't matter if the call was bogus or if it was a tech, all that matters is that he hit his FTs. The game:
http://www.basketball-reference.com/boxscores/201311240OKC.html

Or PER is a production stat. Like CP3 has a higher PER than Magic Johnson. Or a guy like Brook Lopez had a better PER than Kobe last year. It doesn't mean that they are better players or anything like that. Or they impact the game less.

So, advance stats is just something you have to take within context imo. I think you can use the basic stats, and then use advance stats to explain other things.

[OldSkoolball#52]

Considering that Kobe has a higher PER than Larry Bird and Magic Johnson, no he isn't the most overrated.

Not in the playoffs :D


It's widely acknowledged that PER crosses eras very poorly. Paul, Durant, Duncan, Nowitzki, Howard, and McGrady are all ahead of Kareem in playoff PER.

The game has changed way too much to try and make stats uniformly meaningful across eras.


However, there are 9 players who are currently ACTIVE with a playoff PER higher than Kobe's. Now, go ahead and make the case that Kobe's impact is 'intangible'. He's never been a defensive stopper, the proof is in his defensive assignments over the years. Guards Raja Bell and Thabo Sefalosha so he can save his energy for offense. In the finals against the Pistons he got lit up by Rip Hamilton. He also is statistically the most unclutch star in history at the end of games. He shoots poorly in the Finals.



Look, PER is extremely far from perfect. But if it's between PER and "rings" to try and accurately gauge Kobe as a playoff performer? PER at least removes the crutch of team accomplishments from the equation, and represents something closer to what his ACTUAL significance is.

[TheMarkMadsen]

Not in the playoffs :D


It's widely acknowledged that PER crosses eras very poorly. Paul, Durant, Duncan, Nowitzki, Howard, and McGrady are all ahead of Kareem in playoff PER.

The game has changed way too much to try and make stats uniformly meaningful across eras.


However, there are 9 players who are currently ACTIVE with a playoff PER higher than Kobe's. Now, go ahead and make the case that Kobe's impact is 'intangible'. He's never been a defensive stopper, the proof is in his defensive assignments over the years. Guards Raja Bell and Thabo Sefalosha so he can save his energy for offense. In the finals against the Pistons he got lit up by Rip Hamilton. He also is statistically the most unclutch star in history at the end of games. He shoots poorly in the Finals.



Look, PER is extremely far from perfect. But if it's between PER and "rings" to try and accurately gauge Kobe as a playoff performer? PER at least removes the crutch of team accomplishments from the equation, and represents something closer to what his ACTUAL significance is.


Larry has a lower playof per than both..

Kobes 09 playoff run is the highest out of all 3..

[tpols]

Well here's the thing: You wouldn't want to use it for guys who are very dissimilar.

But you know how people say "Dude, Jordan and Horry's rings aren't the same. You can't use rings to compare one guy who's "the man" with another who's not "the man"!!!

It's pretty much like that. You're not gonna compare a defensive players PER to a scorer. But if you're comparing two primarily offensive players and want a snapshot of their effectiveness, PER is going to be as good or better than any simply discussion of points/assists/fg%


I mean how can anyone justify using "rings" to compare similar players, but not PER? Oh, right. Because they prefer the results when you use "rings" instead of PER.

It's no coincidence that Kobe stans are BY FAR the biggest detractors from advanced metrics. Because their hero, their favorite player, is by all analytical accounts the most overrated player in NBA history.
Eh.. As far as offensive players go it is still possible for a lesser impact player to have a better PER. Going back to Kblaze's post, Isiah Thomas could be less efficient and productive than CP3 but the way Isiah took control of games down the stretch and took over when it mattered may have turned the tide of games more than what CP3 did.

Taking over a basketball game whether it be offensively or defensively is the hardest thing to find in a player. It's the most valuable commodity. Timing and sequence of events matter a lot.. And stats don't tell the story of how something happened.

If someone can go off like Lebron in game 6 last year where he literally blocked rebounded stole and scored everything for Miami in a 2 minute stretch to give his team the lead.. Those stats are worth more than if he had accumulated them slowly over a whole quarter. They turned the tide of the game and shocked the opponent. Like a Tyson knockout punch is being compared to a jab.

[OldSkoolball#52]

Eh.. As far as offensive players go it is still possible for a lesser impact player to have a better PER. Going back to Kblaze's post, Isiah Thomas could be less efficient and productive than CP3 but the way Isiah took control of games down the stretch and took over when it mattered may have turned the tide of games more than what CP3 did.

Taking over a basketball game whether it be offensively or defensively is the hardest thing to find in a player. It's the most valuable commodity. Timing and sequence of events matter a lot.. And stats don't tell the story of how something happened.

If someone can go off like Lebron in game 6 last year where he literally blocked rebounded stole and scored everything for Miami in a 2 minute stretch to give his team the lead.. Those stats are worth more than if he had accumulated them slowly over a whole quarter. They turned the tide of the game and shocked the opponent. Like a Tyson knockout punch is being compared to a jab.



I agree.


Basically I just made this thread because there are people here who will use raw stats to discuss and compare players, when it's convenient to their agenda. But then another person uses PER and they flip out.

It's hypocritical IMO. PER is just a collection of stats. If you use stats to make an argument and then say GTFO with PER.... Wtf??

[Deuce Bigalow]

Not in the playoffs :D


It's widely acknowledged that PER crosses eras very poorly. Paul, Durant, Duncan, Nowitzki, Howard, and McGrady are all ahead of Kareem in playoff PER.

The game has changed way too much to try and make stats uniformly meaningful across eras.


However, there are 9 players who are currently ACTIVE with a playoff PER higher than Kobe's. Now, go ahead and make the case that Kobe's impact is 'intangible'. He's never been a defensive stopper, the proof is in his defensive assignments over the years. Guards Raja Bell and Thabo Sefalosha so he can save his energy for offense. In the finals against the Pistons he got lit up by Rip Hamilton. He also is statistically the most unclutch star in history at the end of games. He shoots poorly in the Finals.



Look, PER is extremely far from perfect. But if it's between PER and "rings" to try and accurately gauge Kobe as a playoff performer? PER at least removes the crutch of team accomplishments from the equation, and represents something closer to what his ACTUAL significance is.
Also means that 10 players would have a higher playoff PER than Larry Bird if he was active. So in the end what does that tell you?

Kobe's first 2 seasons brings his averages down. Minus his first 2 playoffs and Magic's last playoffs, Kobe's PER is 22.8, Magic 23.0.

Peak
Kobe: 26.8 ('09)
Magic: 26.2 ('87)

Magic's 5 highest PER average is 24.4
Kobe's 5th highest PER is 24.4

Playoffs with over 25 PER
Kobe: 3 (2 rings, the other a Finals appearance)
Magic: 2 (1 ring)

[Kblaze8855]

It's hypocritical IMO. PER is just a collection of stats. If you use stats to make an argument and then say GTFO with PER.... Wtf??

You really dont see how saying:

"Karl Malone had 40......"

isnt the same as:



All calculations begin with what I am calling unadjusted PER (uPER). The formula is:
uPER = (1 / MP) *
[ 3P
+ (2/3) * AST
+ (2 - factor * (team_AST / team_FG)) * FG
+ (FT *0.5 * (1 + (1 - (team_AST / team_FG)) + (2/3) * (team_AST / team_FG)))
- VOP * TOV
- VOP * DRB% * (FGA - FG)
- VOP * 0.44 * (0.44 + (0.56 * DRB%)) * (FTA - FT)
+ VOP * (1 - DRB%) * (TRB - ORB)
+ VOP * DRB% * ORB
+ VOP * STL
+ VOP * DRB% * BLK
- PF * ((lg_FT / lg_PF) - 0.44 * (lg_FTA / lg_PF) * VOP) ]

Most of the terms in the formula above should be clear, but let me define the less obvious ones:
factor = (2 / 3) - (0.5 * (lg_AST / lg_FG)) / (2 * (lg_FG / lg_FT))
VOP = lg_PTS / (lg_FGA - lg_ORB + lg_TOV + 0.44 * lg_FTA)
DRB% = (lg_TRB - lg_ORB) / lg_TRB

I am not going to go into details about what each component of the PER is measuring; that's why John writes and sells books.

Problems arise for seasons prior to 1979-80:
1979-80 — debut of 3-point shot in NBA
1977-78 — player turnovers first recorded in NBA
1973-74 — player offensive rebounds, steals, and blocked shots first recorded in NBA

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.

Some of these solutions may not be elegant, but I think they are reasonable. After uPER is calculated, an adjustment must be made for the team's pace. The pace adjustment is:
pace adjustment = lg_Pace / team_Pace

League and team pace factors cannot be computed for seasons prior to 1973-74, so I estimate the above using:
estimated pace adjustment = 2 * lg_PPG / (team_PPG + opp_PPG)

For all seasons where actual pace adjustments can be computed, the root mean square error of the estimates is 0.01967.

Now the pace adjustment is made to uPER (I will call this aPER):
aPER = (pace adjustment) * uPER

The final step is to standardize aPER. First, calculate league average aPER (lg_aPER) using player minutes played as the weights. Then, do the following:
PER = aPER * (15 / lg_aPER)

The step above sets the league average to 15 for all seasons.


Just because both use a number....you dont understand the difference?

If I say a guy had 40 and took over down the stretch he had 40 and took over.

If you come in and start with the root mean square error of pace adjustments and giving me his numbers once you factor in lg_PTS / (lg_FGA - lg_ORB + lg_TOV + 0.44 * lg_FTA to get the VOP...


Im talking about basketball.

You...are giving me some bullshit I dont care about. You are not talking about what happened.

You are giving me a number based on what someone decided to value the basic numbers we have which just account for what happened that could be measured.

"Michael Redd scored 26 a game last month...."

Is not the same as PER. Just because both use a number...does not mean someone should use both.

You really dont get how someone can like something as it is....but feel its awful when mixed with 20 other things?

Would you like your favorite food if I mixed it in a blender with your second through fifth favorite food?

No?

Why not?

Its still in there.....

[OldSkoolball#52]

You really dont see how saying:

"Karl Malone had 40......"

isnt the same as:





Just because both use a number....you dont understand the difference?

If I say a guy had 40 and took over down the stretch he had 40 and took over.

If you come in and start with the root mean square error of pace adjustments and giving me his numbers once you factor in lg_PTS / (lg_FGA - lg_ORB + lg_TOV + 0.44 * lg_FTA to get the VOP...


Im talking about basketball.

You...are giving me some bullshit I dont care about. You are not talking about what happened.

You are giving me a number based on what someone decided to value the basic numbers we have which just account for what happened that could be measured.

"Michael Redd scored 26 a game last month...."

Is not the same as PER. Just because both use a number...does not mean someone should use both.

You really dont get how someone can like something as it is....but feel its awful when mixed with 20 other things?

Would you like your favorite food if I mixed it in a blender with your second through fifth favorite food?

No?

Why not?

Its still in there.....



You wouldn't make a very good scientist or mathematician.

But you'd probably make a pretty good priest.


"Sounds good" > Facts :confusedshrug:

[Kblaze8855]

If I were trying to advance the understanding of string theory id give a damn.

Ive been a basketbal lfan all my life. From arguing like the dumb kid I was about Doctor J being better than Elgin Baylor with my uncle and stepfather who watched him to watching replays of games I missed on ballstreams right now.

You people trying to turn a baseline jumper into a quadratic equation just dont strike me as the kind of people id like to talk sports with.

I wonder if I sounded like you in the 80s spitting stats at my uncle about shit I had no clue about.

I was lucky to have him and an old player ive mentioned(Clyde Mayes...played for the Blazers and Pacers...we speak now and then) point me in the right direction.

Id love to get a hold of Clyde and talk to him about Wes Unselds PER rating(14.4) saying he was below average when he led his team to a title and was one of the great leaders and professionals in history.

I bet hed hang up on me. And he should.

Id be acting like an idiot.

[atljonesbro]

If I were trying to advance the understanding of string theory id give a damn.

Ive been a basketbal lfan all my life. From arguing like the dumb kid I was about Doctor J being better than Elgin Baylor with my uncle and stepfather who watched him to watching replays of games I missed on ballstreams right now.

You people trying to turn a baseline jumper into a quadratic equation just dont strike me as the kind of people id like to talk sports with.

I wonder if I sounded like you in the 80s spitting stats at my uncle about shit I had no clue about.

I was lucky to have him and an old player ive mentioned(Clyde Mayes...played for the Blazers and Pacers...we speak now and then) point me in the right direction.

Id love to get a hold of Clyde and talk to him about Wes Unselds PER rating(14.4) saying he was below average when he led his team to a title and was one of the great leaders and professionals in history.

I bet hed hang up on me. And he should.

Id be acting like an idiot.
"Back in my day....."

[Kblaze8855]

All days are mine.

[OldSkoolball#52]

If I were trying to advance the understanding of string theory id give a damn.

Ive been a basketbal lfan all my life. From arguing like the dumb kid I was about Doctor J being better than Elgin Baylor with my uncle and stepfather who watched him to watching replays of games I missed on ballstreams right now.



Ah. So your aversion to advanced metrics is really more about comfort than logic. That's what I figured. But don't worry, that's how it usually is with you old fogies :lol

[GOBB]

If I were trying to advance the understanding of string theory id give a damn.

Ive been a basketbal lfan all my life. From arguing like the dumb kid I was about Doctor J being better than Elgin Baylor with my uncle and stepfather who watched him to watching replays of games I missed on ballstreams right now.

You people trying to turn a baseline jumper into a quadratic equation just dont strike me as the kind of people id like to talk sports with.

I wonder if I sounded like you in the 80s spitting stats at my uncle about shit I had no clue about.

I was lucky to have him and an old player ive mentioned(Clyde Mayes...played for the Blazers and Pacers...we speak now and then) point me in the right direction.

Id love to get a hold of Clyde and talk to him about Wes Unselds PER rating(14.4) saying he was below average when he led his team to a title and was one of the great leaders and professionals in history.

I bet hed hang up on me. And he should.

Id be acting like an idiot.

:roll:

[OldSkoolball#52]

Of course, some old people act more like they're 10 year olds than fogies...

[Legends66NBA7]

A real NBA Fan who watches games: "He played great, 36pts 7reb 7ast 2stls 2blk 47%fg, especially in that 3rd quarter when the opposing team had the momentum, he hit 3 straight shots that pretty much stopped the others team run, so his team will enter the 4th up by 12 instead of up by 4". (This players team hits a 3 to start the 4th, teams up 15, and essentially opens up the game from there and wins in blowout fashion).....This fan watched the game, understood the context of the outcome, and appreciated the Original stats of the NBA (Pts, Reb, Ast)

How stat nerds cover games:

They don't watch it....A week goes by, they see a players stats: "wow, this guy had 25pts 7reb 12ast 57%fg shot the ball 12 times, PER at 32, winshare swag, him>>>the other guy"....Except the Real NBA fan watched that game, and saw Mr Winshares swag, stop shooting to protect a fg%, and then picked a quarter to only pass and get assist which occured in mostly garbage time due to his team building a lead while he was on the bench.


Fact/

Um... again, you can't call anybody a nerd if you reference stats yourself.

[Kblaze8855]

Ah. So your aversion to advanced metrics is really more about comfort than logic. That's what I figured. But don't worry, that's how it usually is with you old fogies


If you find considering defense, leadership, proper playcalling, clutch play, and the dozen other things PER cant measure an issue of comfort being valued over logic...fine. Id consider it....caring about basketball. But old people might just be that way. And I smiled reading this earlier:



. If you think he was better based on game, fine. State your case when the issue comes up. But using "rings" is a complete cop out. It's the falsest of false arguments. It's designed for its simplicity, so that the average rube can use it and put in his 2 cents. Actual analysis excludes too many people, so "rings" is something for the simpletons. Those who actually prefer intelligent discussion laugh at the mere idea of it.

Me 4 years ago:



Per cant account for real defense, ability to score(not points or shooting percentage), clutch play, effort, ball handling(not turnovers...at all), off the ball movement, ability to feed the post, reaction to the double, ability to pass to your man where he wants it to shoot from,willingness to play through adversity or injury, ability to use your off hand, altering shots, ball denial, chemistry with your teammates, or shot selection. To put it simply...PER doesnt measure basketball. It measures the tiny tiny portion of it we have decided to give a numerical value.

Long story short? PER gives people who dont know anything a concrete number to get them out of explaining themselves and it gives people who do know what they are talking about a suggestive but hardly error proof method to use when they are too lazy to explain themselves.

On a basic level...PER like all other formulas exists because people are both lazy and prone to argue. An all knowing formula spares fans the effort of thinking and reason while also giving them an inarguable number to throw out because guys with a string of quadratic equations are hard to argue with and you have to argue the validity of the formula and not if the person using it knows anything about the game.

PER protects the ignorant from the responsibility of explaining their reasoning and protects the rest from having to explain themselves to the know nothings who would otherwise be left furrowing their brows in a vain attempt to understand the situation.

You rely heavily on these numbers you are either too stupid to explain yourself or too lazy/busy to explain yourself. So bottomline?

It like all stats...its just a time saver.


And that was when it was barely known. As time goes on more and more people who couldnt tell you the first thing about the defense Jerry Tarkanians teams played when they had Greg Anthony or how Charles Oakley suddenly became the best alley oop passer in the league...start throwing out strings of advanced stats because they couldnt explain themselves in basketball terms to save their lives.

And it always breaks down to an argument over the validity of the formula itself....so no basketball gets discussed at all.

And we are left talking about how in the **** someone decided to include the .44 for a FTA in true shooting and how that isnt just pulled directly out of someones ass like all the rest of this nonsense.

And it goes into "Oh but you see...the .44 is what Dr.Suchandsuch estimates a blah blah blah is worth so he just decided on that. But its 4.75 in college because you see...."

As if someone just deciding how much to count anything doesnt make it as subjective as opinions are when you dont use the numbers to begin with.

Its just...bullshit.

And for the most part....its bullshit used to duck having to explain with terms that can be argued against by someone with a better understanding of the game.

The object is to make it all objective. Because opinions leave too much room to *gasp*.....talk about the game.

We need something we can prove. Cant argue with a number. Only discredit the means of arriving at it. And so here we are. Bet its 3 years till a claim that ___ is just better than ____ is met with:

"Lol....hes BETTER? Thats an opinion! You cant have those! You cant prove it can you? Then it isnt objective. There is no "better". There is only the data. What you think is too subjective. Tell me what you can prove!"

Actually...my bad...this was said to me shortly after the post I quoted for you:



The point is, your "my opinion blah blah blah" argument is so weak it's not even funny. All it takes is another contrary opinion (which will pop up), and its' gone.


By some idiot who believes Gary Payton is top 25.

No matter how low the bar....the number crunchers will limbo under it.

Bunch of people coming onto a basketball forum trying to eliminate opinion....as if exactly how good someone is could ever be proven anyway.....

Its just ice skating up a hill. And sadly im right next to you. Same hill...different goals.

You are trying to reach a world where matters of opinion can be proven....im trying to reach one where nobody would try to prove it to begin with as much as explain themselves as best they can then chill and watch a game.

But we are both stuck here. You will have to accept opinion just as I will have to accept that the basketball fans I spent my life with dont seem to exist on the internet so I can stop expecting people to act like they do offline in normal conversations.

We both probably need to just take off the ice skates....