? about PER

[pipfan]

I know I could read Hollinger, but I am lazy.

When I read about LBJ's "historic" PER this season (I guess he was on a pace to eclipse MJ's best season) I wondered-isn't PER a per/year stat, based on a league average for that season?

I looked at his raw stats, and they are clearly great-but are they better than his Clev days? I know his D has improved, but he is not the DPOY (has to be Chandler). What am I missing on his "Historic" season?

I hate LBJ-I'll admit. It's embarrasing that I feel that way, but I can't stand the buy. I'll also admit it's a joke to argue anyone other than him for this year's MVP-Durant and Paul have been great, but come on. LBJ should be a 95% 1st place winner. However, how is this SO great a season, when play is clearly watered down due to the Lock Out and the compressed schedule.
Great Season-yes
MVP-Obviously
ALL Time Great Regular Season-I don't see it

[6_Rings]

Ring=no

[illiance]

The league average PER is always 15.

[Mr Grant Hill]

His PER is down already, 30.6, would be 12th best of all time.

[pipfan]

But that league average is for that season only? So in a better year for quality of play, a 15 PER would not be the same as this season, correct?

[SideshowBob]

When I read about LBJ's "historic" PER this season (I guess he was on a pace to eclipse MJ's best season) I wondered-isn't PER a per/year stat, based on a league average for that season?

It is indeed. But the league average is set to 15 every single year, and there is no evidence to suggest that the quality of a "league average" player is any different than it was in Jordan's era

I looked at his raw stats, and they are clearly great-but are they better than his Clev days? I know his D has improved, but he is not the DPOY (has to be Chandler). What am I missing on his "Historic" season?

Several things to get into here. First of all, they aren't anywhere near his Cleveland days. Earlier, when he was going for the record, his per minute scoring and scoring efficiency were at career highs, while his minutes were at a career low. His assist rate was a bit higher than it is now as well, and that was enough to make the difference (He was at just about 33 before, as opposed to just under 31 now). In reality, he wasn't playing nearly at the level that he maintained in his last 2-3 years in Cleveland however, he was only playing better "within the PER system."

Next, his defense hasn't improved. If anything, he was better defensively in 2009 and 10. He's just getting much more media attention for it now. He carried a much larger defensive load in Cleveland (and slightly more effectively I should add).

ALL Time Great Regular Season-I don't see it

While I agree with your sentiment regarding how great of a season this is, I need to point this out. His season "statistically" is heavily deflated by playing with Wade/Bosh. No matter which direction you want to cut it, he's sacrificing usage by playing with them. I've been posting his numbers in games and minutes where Wade and Bosh don't play. Over the last two seasons, we have a sample size large enough for an entire 82 game season, and those numbers DO suggest that had he not switched teams, he would've continued playing at a true "all-time level" (as he did in 09/10) and he probably would have broken the PER record (not that that means much).

[Atmanne]

Even though I still see certain analysts use it, the "he's having a historic season" narrative is about a month outdated.

[mysticbb]

But that league average is for that season only? So in a better year for quality of play, a 15 PER would not be the same as this season, correct?

That is correct, but the quality of play is in average not so much different anyway. We are talking about a difference of about of 2.5% (assuming all the drop is caused by lower quality of play and not by rule changes) in comparison to last season.

The good thing about an adjustment for league average is, that it covers changes in terms of playing level due to rule changes completely. The result has to be read as "above league average" for each season and in this way we can compare players from different seasons with league average adjusted stats.


His season is not so great anymore. He started out really good while everyone else was still in some sort of a funk. And while everyone else picked it up over the course of the season, James even slipped a bit.

Btw, Tyson Chandler is NOT the DPOY of this season. The Knicks are playing better defense without him as they play with him. The biggest part of the defensive improvement is due to better perimeter defense (more forced turnovers, less foul calls, lower perimeter shooting efficiency of the opponents and somehow lower free throw percentage of the opponents). If Chandler would have been the cause for that, those things wouldn't show up when Chandler isn't on the court.
And, the Dallas Mavericks in this season are playing better defense when Nowitzki is on the court than they played last season with Nowitzki/Chandler or Chandler alone.

Chandler, while being a really good defender, is getting overrated due to circumstantial fairy tales.

[parapooper]

His PER and WS/48 went down to 12th and 11th all time when he got banged up for a while but that still means that only KAJ, MJ, Wilt and himself have had better regular seasons (Shaq 2000 and DRob 94 are about tied)
Durant´s season is beyond 100th of all time - and he does not have the intangibles to make up for that.

[SideshowBob]

Pace across eras must also be considered.

Lebron's career statline, pace adjusted to Jordan's era turns into 30.7/8.0/7.7. Pace adjusted to Jordan's Bulls teams, its 30.0/7.8/7.5. That adjustment lacks context (which could favor or hurt the player in question), however you get the idea.

To put into perspective what Lebron was doing earlier when his PER was at 33+, here's a quick adjustment.

Numbers from first 21 games of the season:

Code: Select all

G    MP    PTS    TRB    AST    STL     BLK    TS%    ORTG   GmSc
21   37.5  29.7   8.3    7.0    1.8     0.7   .625    118    24.7

Now here's that statline adjusted to Jordan's first threepeat Bulls (slower paced than late 80s Bulls)

Code: Select all

G    MP    PTS    TRB    AST    STL     BLK    TS%    ORTG   GmSc
21   37.5  30.9   8.6    7.3    1.9     0.7   .625    118    25.7

And now adjusted to the early 90s league average

Code: Select all

G    MP    PTS    TRB    AST    STL     BLK    TS%    ORTG   GmSc
21   37.5  31.6   8.8    7.5    1.9     0.7   .625    118    26.3

Looks a bit stronger now doesn't it? 32/9/8 63%TS is a pretty remarkable season. Of course he fell off that pace, so the point is moot. But it sheds some light as to why it was being held in such high regard.

[Doctor MJ]

But that league average is for that season only? So in a better year for quality of play, a 15 PER would not be the same as this season, correct?

That is correct, but the quality of play is in average not so much different anyway.

In general, I wholeheartedly agree with this assessment. The quality of superstars has a good amount of variability from year to year, but the idea that the 137th best guy from one year to another will be drastically different is silly.

However...

It is my belief that in the last two lockout seasons, offenses have been worse because the offenses have simply played bad across the board. Offenses really do seem to need time to work well together, and until they do, they play worse. So in that sense, were a player to hold the all-time PER record in a year like this, I would see it as having a small asterisk next to it.

[Chicago76]

^ Exactly what I was going to say.

With offenses struggling to sync up early on and perimeter shooting not what it likely would have otherwise been, it also makes sense to put the ball in the hands of your best player and let him take off with it.

The ideal candidate for an exceptional PER season in a strike-shortened season would be a perimeter player (because they can receive the ball easily) who also happens to be able to penetrate or play inside a bit more if his shot isn't falling + create off the dribble: James, Jordan, Erving, Wade are probably the four best candidates over the last 40 years.

[GreenHat]

But that league average is for that season only? So in a better year for quality of play, a 15 PER would not be the same as this season, correct?

That is correct, but the quality of play is in average not so much different anyway.

In general, I wholeheartedly agree with this assessment. The quality of superstars has a good amount of variability from year to year, but the idea that the 137th best guy from one year to another will be drastically different is silly.

However...

It is my belief that in the last two lockout seasons, offenses have been worse because the offenses have simply played bad across the board. Offenses really do seem to need time to work well together, and until they do, they play worse. So in that sense, were a player to hold the all-time PER record in a year like this, I would see it as having a small asterisk next to it.

Offenses have struggled in this lockout season but wouldn't we expect that same decrease for everyone?

The Heat offense was down too so that should be less point and assists for Lebron as well. Is there any reason why he would be immune to the league wide effect?

We don't put an asterisk next to Russell's defensive numbers even though offenses really struggled then, much more than they did this year.

Unless there is a specific reason why one person would benefit over others I don't see a reason for any kind of asterisk.

Of course he fell off the pace so its a moot point.

[Chicago76]

Offenses have struggled in this lockout season but wouldn't we expect that same decrease for everyone?

The Heat offense was down too so that should be less point and assists for Lebron as well. Is there any reason why he would be immune to the league wide effect?

The decrease wouldn't be the same for everyone, because not all offenses are built the same. An offense without a primary creator requires more timing and chemistry to get an open shot than one with a player or two who can get a shot at will. The Heat struggled like everyone else, but the difference is, they could just give the ball to James and ask him to go to work while their offense developed chemistry. More touches and more shots (as long as his FG% wasn't absolutely awful) = higher PER.

Two other sports analogies: imagine a strike shortened NFL season. Team A has a timing based passing attack, while team B has an absolute beast of a RB who can get you almost 5 yards per carry. Team B will do better out of the gate relative to the teams that are more reliant on timing in a passing game.

Baseball regular season vs. post season. Pitchers get saved over the course of the year for later stretches. Pitching gets much more important in the postseasons when rotations shorten, giving more time on the mound for your ace.

Some things may affect all teams to some degree, but they will never affect all teams to the same degree.