GREAT NEWS ABOUT JOHN HOLLINGER!

John Hollinger will be ending his career as a columnist and pseudo-statistician for ESPN and is joining the Memphis Grizzlies' front office as VP-Basketball Operations.

He's going to upgrade their analytics.

HAHAHAHAHAHAHAHAHAHAHAHAHAHAHA!!!!!!

Good riddance to a columnist with a literary PER of 24.7. Great on offense, lousy on defense, but the number only shows half the story.


bob


.

shoot,, if only he was gonna go work for the Heat...

I was reading last night from a number of different sources that Hollinger is one of the most respected statisticians, editors and writers that cover the NBA. His PER and The Hollinger Game Score Formula that he developed are good guidelines when you use them in the right context. They don't take into consideration all aspects of what a player does on the court, but the end result of the stats seem to match closely how a player performs on the court.

Bob, You're joking right? This is an early April Fool's prank played by the Grizz front office on the fans , right? Hollinger giving advice to any NBA team about anything is a joke.

Hi,
It's not a joke. Here is the link.
Oh well, now he's going to sell snake oil to Grizz's ownership...

AK

You can thank Billy Beane for this.

tjmakz wrote:I was reading last night from a number of different sources that Hollinger is one of the most respected statisticians, editors and writers that cover the NBA. His PER and The Hollinger Game Score Formula that he developed are good guidelines when you use them in the right context. They don't take into consideration all aspects of what a player does on the court, but the end result of the stats seem to match closely how a player performs on the court.

I'm not a statistician, but some of my best friends on this forum are statisticians (okay, one). Our in-house statistician (Sam, who makes a living at it) may disagree with you.

From my position as a lay (non-statistician) person, I don't think much of Hollinger. He's like those BCS computer rankings that you look at and go, "Huh?" Yes, the cream generally rises to the top, but an overreliance on a skewed and incomplete set of stats results in curiously out of place results that make you question the value of the rating system. For example, in his NBA rankings (based purely on his selection of stats), the Lakers are ranked 11th out of 30 teams despite having a 9-14 record and being 3-7 in their last 10 games (which his system supposedly puts extra emphasis on). And that puts them one place ahead of the Warriors, who are 15-7 overall and 8-2 in their last 10 games. Maybe the Warriors current record will be an anomaly and the Lakers will ultimately have a better record, but there's no denying that the Warriors are playing far better basketball than the Lakers thus far, and you'd think that any objective statistical measures would reflect that. Ranking teams solely by wins and losses is a better system than what Hollinger does.

As a writer, I don't know much about him because most of his stuff is posted on the "In" (pay) portion of the ESPN site, but from what I've seen, he finds some statistical findings and then writes a column based on that. Well, if you want to make a contention that Miami is the best team, you can find statistics to support that, and if you want to say that they're a seriously flawed team, you can find statistics to support that. My problem is that I think statistics should support a cogent argument, but with Hollinger, statistics are the argument.

My two cents.

To anyone:

There's not much to say about Hollinger that I haven't said before. First of all, he's not a trained statistician. Second, he takes advantage of the fact that numbers automatically are accorded implicit credibility by most fans, whether or not those numbers have relevance and/or context. Third, PER is an offensive calculation only, and a selective offensive calculation at that, and players named Russell, Cousy, Havlicek, S. Jones, Rondo, Worthy, and especially that well-known bum of all time, Michael Cooper (who couldn't even average a 13 lifetime PER), need not apply.

I'm sure everyone believes that Wilt Chamberlain was 38% better than Bill Russell (26.1 PER to 18.9 PER respectively)—right? Riiiiiight. And before anyone comments that there weren't so many stats available in the old days (which, by the way, doesn't stop Hollinger from estimating how the old players WOULD HAVE fared on the missing stats and plugging that fabricated crap into his PER formula), do you really believe Chris Paul is 47% better than Rajon Rondo (25.4 and 17.3 PER respectively)? I believe those two guys are playing currently, right?

Hollinger's PER statistic favors SCORERS—preferably those who have least scoring "competition" from their own teammates. That bias conveniently fits right into the league's marketing of stars over teams. In fact, I'd be willing to bet Hollinger thinks the word "team" Hollinger makes a living by forcing commonly available stats into elaborate formulas, fabricating his own estimates when the stats are unavailable, giving short shrift to matters of context, and capitalizing the public's love affair with figures (not the Marilyn Monroe kind either). I couldn't care less how many of his peers love him. That's simply a reflection on the judgment of the peers.

John Hollinger sucks so badly that his cheeks must be permanently hollowed out and his lips pursed in perpetuity. I wish him luck with the Griz (notice that I didn't specify which kind of luck); but, more to the point, I wish the Griz good luck with him. Wait until the first time they realize how much of an acquired player's performance depends on his chemistry and role with his previous team(s).

Sam

If Hollinger had cast judgment on naval architecture, he would probably say the Titanic was the best built ship of the 20th century.

sam wrote:
John Hollinger sucks so badly that his cheeks must be permanently hollowed out and his lips pursed in perpetuity.

Sam

Now that paints a picture worth a thousand words!

All-star game voting, MVP voting, Playoff MVP voting, Hall of Fame voting and no other award is based on a players PER. I don't understand the hostility toward Hollinger's efforts to bring another angle toward stats in the NBA. He has always recognized that there are obvious limitations to his PER. It does not measure when you help on defense, it does not measure setting picks for others and doesn't measure diving to the floor for a loose ball. As I said previously, you use his stats (if you want) as a guideline. Are we angry that his stats slight players that we like or are on our team? There are so many new stats that John and others have developed such as True Shooting Percentage that teams are now really paying attention to.

Here are some interesting points from Hollinger's farewell column.


Eventually, however, the best information wins. And ultimately, analytics is about providing more and better information to supplement what is already available. Steadily, that realization has taken hold. Not only is it now acceptable to mention things like true shooting percentage in polite conversation, but there's also been a real quantum shift in front offices like the one I'm about to join.




It's hard to believe this is true, but just eight years ago very few teams showed any interest in analytics, and those who did wouldn't admit it publicly. Seriously. Teams employed analytics people they wouldn't even mention in their directory for fear of ridicule.



In less than a decade, teams have reversed course: Now, if anything, many try to promote how much they're doing with analytics. At least two-thirds of the league's teams have invested in this area, and while a few of them are just checking a box, most are seriously committed to it.

http://espn.go.com/nba/story/_/page/PERDiem-121214/john-hollinger-farewell-column

Do fans have the same hostility toward the developers of WAR in baseball?

Also, the next time anyone wants to use the NBA Trade Machine, you can think of John Hollinger because I heard that he is the one that developed and maintained that website.

TJ,

I can't speak for others, but my own hostility is aimed at a guy who has succeeded (and profited from) foisting what I call "pop stats" off on the public by using the excuse that they're better than nothing. Maybe they are, and maybe they aren't; but I happen to believe that I've identified enough serious flaws in much of his work over the years to err in the direction of "maybe they aren't."

And yet, even though giving lip service to that suspicion, so many people use a stat like PER as though it were gospel—just as they use the ± to evaluate individual players (and I'm NOT blaming that on Hollinger....it's just an example of the sanguine misuse of stats).

I'm not sure exactly what genius Hollinger displayed by developing the NBA Trade Machine (if he really did). It seems to me that it's no particular statistical advance—simply committing to writing what general managers and experts like Jan Volk have done in their heads for many years.

If people are willing to believe that Chris Paul is 47% better than Rondo (and that's exactly the impression that the public receives from the PER stat because disclaimers are not cited every time PER is referenced), good luck to them.

Anyway, it seems petty even to discuss something like Hollinger when so many more important and egregious things are happening in the world. It's just a reflex reaction I have because I've devoted so much of my life to trying to use stats correctly, and I view it as my right to take anything less than that effort personally.

Sam