Current PER #s

[illmatic99]

nearly halfway through the season, figured these numbers have some sort of validity now.

Dame
12-13: 16.5
13-14: 19.9

Wes
12-13: 14.1
13-14: 17.7

Nic
12-13: 15.8
13-14: 16.5

LA
12-13: 20.5
13-14: 22.8

RoLo
12-13: 19.0
13-14: 17.0

Mo
12-13: 14.5
13-14: 12.4

And for good measure, TRob
12-13: 11.0
13-14: 15.1

So outside of Lopez/Mo, everyone else has taken a significant step up. LA has crossed into perpetual all star level and Dame climbs into borderline AS status. I haven't looked too much at other teams but it is crazy how many guys we have >17-- ours has to be the best starting 5 in the NBA.

 

[Blazer Fanatic]

I was going to start another thread about the teams improvement, but this seems like a good spot.

Are you gellin'?

Sochi edition (gold, silver, bronze):

The rebounding, assists, and points per game really stand out. I'd say this team is gellin' quite well.


In wins, Portland shoots over 46% from the field, 41% from 3, and 83% from the FT line.

In losses, they shoot 43%, 33% from 3, and under 80% from the FT line, as well as 2 fewer assists, 1 more turnover, 2 more PFs, and 5 fewer points per game.


Batum has been the game’s high assist man 10 times this season, Lillard 7 times, Mo 5 times, and the opponent 15 times.

Portland is 9-1 when Batum is the game’s high assist man.
Portland is 3-2 when Mo is the game’s high assist man.
Portland is 7-0 when Dame is the game’s high assist man.
Portland is 9-6 when the opponent has the game’s high assist man.

 

[Natebishop3]

It's really surprising to me that Nic doesn't have a higher PER. Why is that?

I'd love to see Dame break 20 and stay there for the rest of the season.

 

[e_blazer]

It's really surprising to me that Nic doesn't have a higher PER. Why is that?

I'd love to see Dame break 20 and stay there for the rest of the season.

I'm certainly not a PER guru, and I'm not even particularly a fan of it as anything other than a rough measuring stick, but I think that Nic gets dinged by Hollinger's formula because he doesn't shoot enough. The Blazers play at a high pace and Hollinger's PER formula adjusts the basic calculation to factor in a team's pace. Since Portland plays at a high pace, but Nic doesn't shoot as often as he might given the number of available possessions, his relatively mid-level scoring numbers get reduced in the adjustment process.

 

[PapaG]

I'm certainly not a PER guru, and I'm not even particularly a fan of it as anything other than a rough measuring stick, but I think that Nic gets dinged by Hollinger's formula because he doesn't shoot enough. The Blazers play at a high pace and Hollinger's PER formula adjusts the basic calculation to factor in a team's pace. Since Portland plays at a high pace, but Nic doesn't shoot as often as he might given the number of available possessions, his relatively mid-level scoring numbers get reduced in the adjustment process.

Nic is also a low-usage guy.

A 16.4 PER on 17 USG is very good for any player.

 

[BlazerCaravan]

He also has a goodly number of turnovers given his higher playmaking responsibility. His TOV% is the highest among the starters, and tied with Joel Freeland. That hurts him, iirc.

 

[Blazer Fanatic]

I'm certainly not a PER guru, and I'm not even particularly a fan of it as anything other than a rough measuring stick, but I think that Nic gets dinged by Hollinger's formula because he doesn't shoot enough. The Blazers play at a high pace and Hollinger's PER formula adjusts the basic calculation to factor in a team's pace. Since Portland plays at a high pace, but Nic doesn't shoot as often as he might given the number of available possessions, his relatively mid-level scoring numbers get reduced in the adjustment process.

Yup, that's part of it, as well as shooting efficiency. Hollinger says he "calibrates" for it by including league average, but it's more akin to putting duck tape on a Dixie cup, still not very useful to describe Batum's contribution. And certainly doesn't explain why DeAndre Jordan has a higher PER.

PER requires a LOT more context than many who cite PER stats seem to realize.

Nic is also a low-usage guy.

A 16.4 PER on 17 USG is very good for any player.

Yup.

Also, of the top 15 SFs in PER, only Iguodala has a higher Assist Ratio (another "superior" stat that Hollinger made up). PER is all about offense, and play making just doesn't stack up with shots and usage in Hollinger's PER.

PER also isn't about defense (stl, blk), nor does it attempt to quantify anything not on a box score, or a player in relation to his team and role. Just have to take it with a grain of salt and realize if this is how you calculate it:

uPER = 1/min ( 3P + [ 2x3 x AST ] + [ ( 2 - factor x tmAST/tmFG) x FG] + [ 0.5 x FT x ( 2 – tmAST/tmFG + 2/3 x tmAST/tmFG ) ] - [ VOP x TO ] - [ VOP x DRBP x ( FGA - FG ) ] - [ VOP x 0.44 (0.44 + ( 0.56 x DRBP) ) x ( FTA - FT) ] + [ VOP (1 - DRBP) x ( TRB - ORB) ] + [ VOP x DRBP x ORB ] + [ VOP x STL] + [ VOP x DRBP x BLK] - [PF x (lgFT/lgPF - 0.44 x lgFTA/lgPF x VOP) ] )

Where:

factor = 2/3 – [ (0.5 x lgAST/lgFG) / (2 x lgFG/lgFT) ]

VOP = lgPTS / [ (lgFGA – lgORB + lgTO + (0.44 x lgFTA) ]

DRBP = (lgTRB – lgORB) – lgTRB

tm = team lg = league DRBP= defensive rebound percentage

Then uPER is adjusted for pace and “normalized” to the league and becomes PER.

PER = [ uPER x ( lgPace/tmPace) ] x 15/lguPER

... and it takes an entire book to defend it, it's got some holes.

PER tends to favor big men, who end possessions by getting defensive rebounds. A point guard (Lillard) who allows a lot of shots won’t be as affected after that giant formula is applied as a center (RoLo). And a SG (Matthews) who harasses the snot out of an opponent and forces them to take a bad shot won’t benefit as much as a PF (Aldridge) who is closer for a rebound.

And then there's the whole, RoLo is a huge reason every starter's rebounding number have gone up, yet that is hardly reflected in his PER.

I could go on for hours... lol

He also has a goodly number of turnovers given his higher playmaking responsibility. His TOV% is the highest among the starters, and tied with Joel Freeland. That hurts him, iirc.

That's a small fraction of it, but doesn't explain why his PER is lower than Michael Beasley.

 

[Blazer Fanatic]

The short answer for RoLo and Batum is, PER doesn't accurately describe them. Does that make it a bad stat. I say yes, but a lot of people don't, so agree to disagree. lol

PER is and individual stat, and basketball is a team sport. If you’ve ever watched the Blazers this year and found yourself cheering for an effort play by one or more Blazers that didn’t result in a box score stat, then we have a problem. No account for Batum’s defense or hockey assist, and no accurate account for his blocks and steals. No accurate picture of RoLo's contribution either. A defensive rebound ends a possession, and a lay-up is 2 points, but RoLo doesn’t get any credit for boxing out or sealing off his man to open the lane. And he does that more often and better than most in the NBA, which requires very unselfish play and a firm understand of his role. He just wants to win, which isn't cliche because he's actually willing to give up the stats that lead to PER to increase the PER of those around him. Every single Blazer starter has increased their RPG, and their PER this season for instance, even Batum. But RoLo's has gone down. PER just isn't an accurate stat unless you just want to look at one's ability to grab defensive rebounds or compare volume shooter.

Reading what goes into these stats is quite mind numbing, and I am inclined to just grab my favorite beverage and enjoy the Blazer win.

 

[Wizard Mentor]

Yup, that's part of it, as well as shooting efficiency. Hollinger says he "calibrates" for it by including league average, but it's more akin to putting duck tape on a Dixie cup, still not very useful to describe Batum's contribution. And certainly doesn't explain why DeAndre Jordan has a higher PER.

PER requires a LOT more context than many who cite PER stats seem to realize.
...
And then there's the whole, RoLo is a huge reason every starter's rebounding number have gone up, yet that is hardly reflected in his PER.

I could go on for hours... lol

That's a small fraction of it, but doesn't explain why his PER is lower than Michael Beasley.

Preach it!

If PER is used as an all encompassing stat, it's TERRIBLE.

OTOH, people often use PPG or RPG as all encompassing stats and PER is much, much, much better than those most of the time.

Most people prefer the "eye test" to any stats and that leads to undeserved derision of stats, even when the "eye test" is obviously flat out wrong (as in the case of "volume" shooters).

 

[magnifier661]

I was going to start another thread about the teams improvement, but this seems like a good spot.

Are you gellin'?

Sochi edition (gold, silver, bronze):

The rebounding, assists, and points per game really stand out. I'd say this team is gellin' quite well.


In wins, Portland shoots over 46% from the field, 41% from 3, and 83% from the FT line.

In losses, they shoot 43%, 33% from 3, and under 80% from the FT line, as well as 2 fewer assists, 1 more turnover, 2 more PFs, and 5 fewer points per game.


Batum has been the game’s high assist man 10 times this season, Lillard 7 times, Mo 5 times, and the opponent 15 times.

Portland is 9-1 when Batum is the game’s high assist man.
Portland is 3-2 when Mo is the game’s high assist man.
Portland is 7-0 when Dame is the game’s high assist man.
Portland is 9-6 when the opponent has the game’s high assist man.

Wow the rebounding, Assists, and Points have really improved!!!! The 51 rebounds per game is the goal I wanted for this team. They do that, then we are in damn good shape!

 

[Natebishop3]

I don't think there's a single stat that will tell the whole story about a player, especially not one like Batum or Lopez. It's just another nice way of tracking progression, but there isn't an end-all, be-all in the stat world. I prefer to lean on the eye test and then back it up with stats if I'm trying to make a point.

 

[Draco]

OTOH, people often use PPG or RPG as all encompassing stats and PER is much, much, much better than those most of the time.

This is why I always thought PER was great. Its a way to quickly make a rough estimation of a players effectiveness. Is it the be all end all? Obviously not. But its far better then just looking at PPG which is what many fans, and the Knicks front office, seems to do.

 

[KeepOnRollin]

This is why I always thought PER was great. Its a way to quickly make a rough estimation of a players effectiveness. Is it the be all end all? Obviously not. But its far better then just looking at PPG which is what many fans, and the Knicks front office, seems to do.

If you take into account minutes played per game then I think PER is perfectly fine in comparing players of a similar position.

 

[mook]

I think it's perfectly fine if you take into account position and role, and provided you understand there's not much difference when you get down between 1 or 2 points. The 15.0 PER guy may be better than the 16.8 guy, but most of the time the guy with the 19 PER is significantly better than the 15 PER if they have the same role on the team. Rolo and Hickson are a great example of an exception, but it's not that common.

 

[Blazer Fanatic]

Wow the rebounding, Assists, and Points have really improved!!!! The 51 rebounds per game is the goal I wanted for this team. They do that, then we are in damn good shape!

Well, I remember this came up pre-season, with Homer's edge D. Deckard saying finishing 1st in rebounding during pre-season didn't mean anything.

You were talking about a RPG number, and me or someone else mentioning it's not really the RPG, but where they rank in relation to the league. But it's funny how it worked out that way. lol

 

[Hobbesarable]

If you take into account minutes played per game then I think PER is perfectly fine in comparing players of a similar position.

PER has its limitations when comparing players offensively and more so defensively.

 

[Blazer Fanatic]

I think it's perfectly fine if you take into account position and role, and provided you understand there's not much difference when you get down between 1 or 2 points. The 15.0 PER guy may be better than the 16.8 guy, but most of the time the guy with the 19 PER is significantly better than the 15 PER if they have the same role on the team. Rolo and Hickson are a great example of an exception, but it's not that common.

But if a stand alone stat requires so much explanation (an entire book) and extra context for anyone not in the top 10-20, doesn't that defeat the purpose of the creating the stat in the first place? It's pretty clear to me that if one's definition of MVP is only about being a stat stuff, then PER is good for predicting the offensive half that goes into MVP consideration. Other than that, it's just a polished turd that most casual fans don't understand what it signifies and, in turn, misuse it to make arguments about a players value to a given team.

I would argue it is very common. And the close you get to 15 PER from either the top or bottom, the less useful the stat is. I just feel there are too many exception for people to put any weight in the value of PER as THE stat. Volume shooters (Gay/DeRozean/Carmelo/Durant), unselfish players (RoLo Batum), stat grabbers (Love, Hickson), playing out of position or being guarded like they are out of position (LA last season, Horford), playing on a bad/good team, being a great/poor defender... these are all exceptions one must account for when considering the merits of PER being an accurate statistic.

 

[Blazer Fanatic]

If you take into account minutes played per game then I think PER is perfectly fine in comparing players of a similar position.

Assuming they all play the same role on their respective teams. Batum's is severely penalized, even though the Blazer are the #1 offense in the NBA because of his usage, and that his assist ratio is ridiculously high as opposed to his scoring stats. Someone has to pass to get all those open shots.

 

[Natebishop3]

Assuming they all play the same role on their respective teams. Batum's is severely penalized, even though the Blazer are the #1 offense in the NBA because of his usage, and that his assist ratio is ridiculously high as opposed to his scoring stats. Someone has to pass to get all those open shots.

Yeah, Nic is essentially our point guard right now.

 

[Blazer Fanatic]

PER has its limitations when comparing players offensively and more so defensively.

I'd take it a step further, and say PER doesn't tell you squat about a players defensive ability. A defensive stand ends with a defensive rebound, like a period at the end of a sentence (defense being the sentence). So, whomever gets that rebound, it isn't a good indicator for who played good defense. Nor is someone who blocks a shot or steals the ball, as the guy who poked the ball away doesn't always recover the ball and get credit for the steal... etc.

In fact, I don't even know why Hollinger even bothered to include steals and blocks. It's like just throwing some more numbers in that might boost PER to determine an MVP, but doesn't add anything in comparing players defensively.

 

[BlazerCaravan]

Yeah, Nic is essentially our point guard right now.

This is a thing I'm totally okay with. Matthews posts up like a SF, spots up like a SF, and is now being played like an SF. Lillard is a combo guard who can rest on the court while Batum sets LMA up. If half our plays are getting a contested pass into our PF, why not let the tall player with great passing skills give him that ball? That sort of play making is low risk and high reward.

 

[Hobbesarable]

Yeah, Nic is essentially our point guard right now.

I used to think a team needed a point guard to run the team. That was conventional thinking when Magic and Stockton (two of the all-time greats) played during the 80's and 90's. However, our team generally goes with Batum and Dame as the playmakers and everything else develops around them.

 

[Wizard Mentor]

You would need at least a chapter of a book to tell how RPG measures a players effectiveness, and the conclusion to that chapter would not be very satisfying. So, should we not talk about RPG? No, we SHOULD talk about RPG, but in context which is also what you should be doing with PER. I'm pretty sure that's what you're saying in some of your posts, correct me if I'm wrong.

But if a stand alone stat requires so much explanation (an entire book) and extra context for anyone not in the top 10-20, doesn't that defeat the purpose of the creating the stat in the first place? It's pretty clear to me that if one's definition of MVP is only about being a stat stuff, then PER is good for predicting the offensive half that goes into MVP consideration. Other than that, it's just a polished turd that most casual fans don't understand what it signifies and, in turn, misuse it to make arguments about a players value to a given team.

I would argue it is very common. And the close you get to 15 PER from either the top or bottom, the less useful the stat is. I just feel there are too many exception for people to put any weight in the value of PER as THE stat. Volume shooters (Gay/DeRozean/Carmelo/Durant), unselfish players (RoLo Batum), stat grabbers (Love, Hickson), playing out of position or being guarded like they are out of position (LA last season, Horford), playing on a bad/good team, being a great/poor defender... these are all exceptions one must account for when considering the merits of PER being an accurate statistic.

 

[Natebishop3]

I used to think a team needed a point guard to run the team. That was conventional thinking when Magic and Stockton (two of the all-time greats) played during the 80's and 90's. However, our team generally goes with Batum and Dame as the playmakers and everything else develops around them.

I think Stotts sees Dame as such a gifted scorer that he leans on Nic to run the offense more often than not, but I also think that Nic might be a more gifted passer than Dame. He has been great at running the pick and roll with Rolo.

 

[Blazer Fanatic]

You would need at least a chapter of a book to tell how RPG measures a players effectiveness, and the conclusion to that chapter would not be very satisfying. So, should we not talk about RPG? No, we SHOULD talk about RPG, but in context which is also what you should be doing with PER. I'm pretty sure that's what you're saying in some of your posts, correct me if I'm wrong.

I see what you're saying. Yes, you absolutely have to use a great deal more care and context when using PER.

To me, RPG is just something that happens in a game. It's one way to end an offensive possession. It's doesn't take a paragraph, let alone a book, to explain and defend it. And, I don't think RPG goes beyond its limitations to describe what happened during a game, or a players performance. PER does in my opinion, in fact ignoring half of the game (defense), and arbitrarily weighting existing stats to try and combine them for comparison. Granted, you can't really create a model for data that doesn't exist and expect it to produce anything meaningful, so Hollinger just made a model with what data existed, threw it in and added weights till it looked like the best player had the highest number. I mean, I know he's genuinely a smart guy, and even a pretty nice guy from the interviews I've heard, but did we really need PER to better describe player contribution? No. It's a search for short hand and lazy imo, and basketball just isn't that simple. In other words, Batum's skill set isn't adequately described in a tweet. PER is... even though it doesn't do that either, but at least it fits within the character limit.


If I make an argument without all the pertinent facts, at the least you'd reserve judgment assuming you'd get them all at some point, and then decide if my argument is represented by the facts.

Hollinger doesn't have all the facts. He only has what is in the box score, which apparently is enough to give the appearance that if you manipulate them all into one number, it can predict... "something."

The issue I've seen most often is fans take PER, and cite it as "Beasley" is better than "Batum" because of "PER". Clearly this is my personal experience, but it happens often enough that I can say with great confidence (in my experience), most fans on message boards don't understand the limitation of PER or what it even represents, and fail to account for context that explains why a player has a higher or lower PER, running counter to whatever Hollinger's model predicted.

 

[Blazer Fanatic]

I think Stotts sees Dame as such a gifted scorer that he leans on Nic to run the offense more often than not, but I also think that Nic might be a more gifted passer than Dame. He has been great at running the pick and roll with Rolo.

Maybe he's better, but I think it's more so that he's taller. Batum's length gives him a distinct advantage to enter the the ball in the post (same reason Nic is the out of bounds guy), and he's harder to double for the same reason. Oh, and, double Batum and leave LA or Lillard open? lol Yes please. It's hard to trap once and recover. It's near impossible to throw a second double team on Batum who is opposite Lillard and not get burned. It's either a pick n role with RoLo or someone else, or the ball gets tossed down to a corner 3, or back to dame for a 3 up top. It's just funny thinking about how crazy hard it is to defend Portland.

 

[riverman]

"Duck tape on a Dixie Cup"...this is the title of my next composition! Thank you Blazer Fanatic! Brilliant! Sounds like the title to a Tom Robbins novel.