Hollinger: The Fluke Rule (mentions TP)

[ElNono]

In case somebody missed it, snagged it from downstairs:

Hollinger expects a sharp decline in Odom's production

Originally Published: November 28, 2011

Odom had a career season at age 31. That's the good news. The bad news is he is one of seven players whose stats can be expected to sharply regress this season, based on a phenomenon called the Fluke Rule. While these seasons are not all flukes per se, the rule of thumb still holds: Any player past the age of 28 whose player efficiency rating (A) jumps by more than three points in a season and (B) is above 14 will see his PER regress the next season, on average, by roughly the same amount. The odds are better than 90 percent that his PER will go down, and in many cases the decline is steep.

This is particularly true for players like Odom that are on the wrong side of 30 and had never played this well previously. Odom's PER last season was a career high because he shot the lights out both in the basket area (70.2 percent) and on 3-pointers (38.2 percent). However, based on his career norms he seems unlikely to sustain such lofty percentages.

http://espn.go.com/photo/2006/1102/pg2_odom_275.jpg

As a reminder, here's how last year's Fluke Rule class fared (see chart). While it rarely works out this neatly across the entire group, the mean Fluke Rule player will lose three points on his PER the following season. (Nice Fluke Rule players are similarly affected).


Fluke Rule, Class of 2009-10
Player 2008-09 PER 2009-10 PER 2010-11 PER Change
Corey Maggette 16.91 20.40 15.89 -4.51
Jamal Crawford 15.15 18.50 14.29 -4.21
Nazr Mohammed 7.71 19.64 15.98 -3.66
Ben Wallace 12.18 15.84 12.29 -3.55
Sam Dalembert 13.22 16.84 14.10 -2.74
Luke Ridnour 12.95 17.81 15.08 -2.73
Carlos Boozer 17.28 21.42 18.90 -2.52


This year, the other seven players joining Odom in Flukeville are Brad Miller, Tony Allen, Chris Andersen, Earl Boykins, Tyson Chandler, Tony Parker and Chris Wilcox. Miller may not play this year after having microfracture over the summer, but of the other seven we can expect six to decline. Parker and Chandler, who were 28 years old on Dec. 31 of last season, are slightly less susceptible than their older cohorts, but all will be hard-pressed to maintain last season's production

Fluke Rule, Class of 2010-11
Player Team 2009-10 PER 2010-11 PER Change
Chris Wilcox Det 12.04 18.09 +6.05
Tyson Chandler Dal 12.58 18.45 +5.87
Earl Boykins Mil 12.98 17.62 +4.64
Tony Allen Mem 14.23 18.40 +4.17
Tony Parker SA 16.49 20.44 +3.95
Lamar Odom LAL 15.98 19.50 +3.52
Brad Miller Hou 12.96 16.37 +3.41
Chris Andersen Den 15.88 19.03 +3.15


While the eight players in this season's crop run the gamut from "no freaking way he does that again" (Wilcox) to "potential glaring exception" (Parker), Odom represents a fairly typical Fluke Rule player and it's clear how his numbers could fall back to earth. Last season he blew away his career highs in field-goal percentage, 3-point percentage and TS%. If those revert back to his usual levels -- which is nearly always what happens -- his PER will come along for the ride. The result would still leave Odom as one of the most effective sixth men in the game because of his defense, rebounding, ballhandling and not insignificant scoring. But it's unrealistic to expect him to shoot 70 percent at the rim and drill 38 percent of his 3s.

John Hollinger is an analyst and writer for ESPN and a fervent member of the CoM

http://insider.espn.go.com/nba/story?page=2011-12-lal-preview&_slug_=losangeles-lakers-player-profiles&action=upsell&appRedirect=http%3a%2f%2finsider.espn.go.com%2fnba %2fstory%3fpage%3d2011-12-lal-preview%26_slug_%3dlosangeles-lakers-player-profiles#reserves

[TDMVPDPOY]

add diaw and any mofo on a contract year

[Bruno]

In addition of only looking at stats to analyze basketball, which is stupid, Hollinger is too a bad statistician.

To determine what he called "fluke year", he just do the PER difference between the year before and this year. The flaw of this method is that the "year before" isn't necessarily representative of the player's level. The proper way to determine these "fluke years" would been to compare their PER to the average PER of the 3 previous years

Parker's case show how Hollinger's method is flawed:
Parker's PER:
05-06: 20.8
06-07: 21.4
07-08: 20.0
08-09: 23.4
09-10: 16.4
10-11: 20.4
The fluke isn't Parker raising by 4 points between 09-10 and 10-11. His 10-11 PER is very similar, or even below, to what he does in his career. The true "fluke" was Parker low PER in 09-10. If you look a little behind statistics, this "fluke" year can easily be explained by injuries.


About PER, it's noticeable that Manu is actually second in the league just behind Lebron. A biggest surprise is that Blair is 7th in the PER ranking. It quite show how both have had a great season start.

[Obstructed_View]

Yeah, Not sure how it escaped him that Parker's fluke year was the season before when he was hobbled.

[timvp]

Yeah, Parker has been a ~20 PER player during his prime years. It was up 15% one year due to injuries to others and down 15% one year due to injuries to himself. There's very little noise there when looking at a bigger picture.

So far this year? 19.81. Unless injuries interfere, TP will land between 20 and 22 again this year. I enjoy Hollinger's work but calling TP's season last year even a "potential glaring exception" is laughably stupid. TP is as predictable when it comes to PER as any player in the NBA.

[Obstructed_View]

I prefer this Tony Parker, who seems good at setting up his teammates while picking his spots to score and not beating himself up. If he can keep his assist to turnover ratio like this and stay healthy the Spurs will be really good.

[ElNono]

No comment on my "fervent member of the CoM" addition?

:depressed

[Cane]

Hollinger's Hollinger, but Parker's playoffs PER slipped, especially compared to his reg season numbers last year. He had a 20.4 PER in the reg season and in the small sample size that was the Grizz series, his numbers dropped to 17.6 PER, which is somewhat significant. Duncan's been slipping as well :depressed

Parker's been playing pretty good so far, but he doesn't seem to be 100% physically either. To me, it looks like he's having more trouble than usual getting to the paint and converting but he has been playing against a lot of length in the paint and also had to play against premier guards like CP3 too.

As long as Parker can continue to hone his court vision and shooting skills, then he can laugh at PER. But unfortunately, Parker is the type of player, a score-first small guard that thrives in the paint, whose decline is likely to be sharp and significant. Would not be that surprising if Parker's PER did drop, but who cares as long as there are W's

[therealtruth]

I prefer this Tony Parker, who seems good at setting up his teammates while picking his spots to score and not beating himself up. If he can keep his assist to turnover ratio like this and stay healthy the Spurs will be really good.

Agree. I like this version of Parker who averages 7+ apg and scores when it makes sense. It also makes it harder for teams to gameplan against him because they have to worry about the other guys. He did the same thing early last year and even was at career high assists. Hopefully he keeps it up because it makes the team more dangerous.

[Man In Black]

In addition of only looking at stats to analyze basketball, which is stupid, Hollinger is too a bad statistician.

To determine what he called "fluke year", he just do the PER difference between the year before and this year. The flaw of this method is that the "year before" isn't necessarily representative of the player's level. The proper way to determine these "fluke years" would been to compare their PER to the average PER of the 3 previous years

Parker's case show how Hollinger's method is flawed:
Parker's PER:
05-06: 20.8
06-07: 21.4
07-08: 20.0
08-09: 23.4
09-10: 16.4
10-11: 20.4
The fluke isn't Parker raising by 4 points between 09-10 and 10-11. His 10-11 PER is very similar, or even below, to what he does in his career. The true "fluke" was Parker low PER in 09-10. If you look a little behind statistics, this "fluke" year can easily be explained by injuries.


About PER, it's noticeable that Manu is actually second in the league just behind Lebron. A biggest surprise is that Blair is 7th in the PER ranking. It quite show how both have had a great season start.
Where is the LIKE button on this here muthasuckin' board?
:hat

[BillMc]

While I agree that Hollinger made a mistake on Tony, I do agree that Lamar is in for an awful year in Dallas.