Dave Berri on John Hollinger

The connection between Wins Produced and PER – and as a result the connection between Dave Berri and John Hollinger — is pretty much as old as the Wages of Wins Journal itself. With John Hollinger joining an NBA front office, the fine folks at Three Shades of Blue (a Memphis Grizzlies Blog) sent Dave a few questions about Hollinger. It’s a fun read and definitely worth checking out:

David Berri Educates Us On John Hollinger

11 Responses to "Dave Berri on John Hollinger"

marparker says:

December 18, 2012 at 3:47 pm

Its getting tired.
Andres Alvarez says:

December 18, 2012 at 3:56 pm

Marparker,
Know you’re a long timer. That said, it’s a lot like Kobe :) Same arguments year after year, but as long as Kobe is playing and shooting a lot, they’ll come up. Hollinger’s still around (was still using PER) and still in basketball news. Way it goes.
steved says:

December 18, 2012 at 5:57 pm

I don’t think PER will form a large part of Hollinger analysis. The flaws in it are well established, especially the shooting efficiency issue, What Hollinger will use we will likely never know as that is something called competitive advantage.

PER to me was always the simple to understand metric for the masses.
Mike says:

December 18, 2012 at 7:12 pm

PER to me was always the simple to understand metric for the masses.

I disagree – PER is one of the most useful advanced metric available, because it predicts salary really, really well. That’s invaluable as a tool to assess players salary expectations and reality (with WP48) and find “bargains”.
aszzzbrokerman says:

December 19, 2012 at 9:52 am

Wins Produced is no.1 that’s why I come here all the time. Time and time again predictions are made based off of WP and they become true more often than not. The record speaks for itself. That being said it’s starting to come off as petty. Dave comes off like a bitter ex in this one. You have done the superior research why lower yourself?
Dave says:

December 19, 2012 at 10:50 am

Just to be clear..
I was asked some questions by Chip Crain, who does not appear to be happy with Hollinger directing his team (primarily because Chip also think WP is a more accurate picture of performance). All I did was answer the questions. I didn’t seek out the interview.

And as I have said before, Hollinger can promote PER or any other model he thinks makes sense as much as he likes. I am not about to spend my time making comment after comment all over the Internet denouncing someone I disagree with. But if I agree to an interview with someone, I will do my best to answer the questions.
Kent Bray says:

December 19, 2012 at 8:58 pm

Has wins produced been given the simple statistical test of projecting how well a team will do when the roser is reshuffled?
Tommy Grand says:

December 20, 2012 at 8:44 am

I don’t think Professor Berri sounded petty at all. I don’t know where you get that. I think this hiring is evidence of intelligence in the Memphis organization. Assuming they do hire some good, traditional scouts, mix traditional scouting with smart statistical analysis, and avoid using PER to rank players in terms of efficiency (3 rational assumptions, I think), there’s no reason to worry and some reason for hope. Of course, within 4 years, all NBA teams will have smart statistical analysts on staff. Many already do. There may be a shortage of 7 footers who can ball, but there’s a surplus smart stat guys and gals (in the US alone) who’d love to quit some accounting firm and go to work for an NBA franchise. Every team will have good analytics soon. The next level is to get superior data.

By the way – in case any NBA owner is reading – I “created” a system that’s superior to WP. By pure coincidence, my unique invention is exactly like Professor’s Berri’s system, except I review individual game film and count charges taken. For each charge a player takes, I credit him one steal. Then I re-compute.
gelfnerd says:

December 20, 2012 at 3:15 pm

If you go back and read Hollinger’s incredible amount of articles it is abundantly clear that PER plays a very minimal role in nearly all of his analysis. When he breaks down offense and defense as well as players on his PER DIEM column the analysis is far more in depth than these PER v. WINS PRODUCED arguments give him credit. Further, NEil Payne makes some very good points regarding the difference between the two.

Just to put the point home, Hollinger isn’t beating vegas projections by relying on PER that much should be abundantly obvious.
Mike G says:

December 21, 2012 at 2:38 pm

Like Tommy above – in case any NBA owner is reading – I ALSO “created” a system that’s superior to WP. By pure coincidence, my unique invention is exactly like Professor’s Berri’s system, except I analyze the facial characteristics of each player.

It turns out that LESS attractive NBA players hook up with groupies more often, to compensate for their insecurities. This means less rest, and lower WP48.

The arbitrage here is simply if you could sign those NBA free agents who are effective and ugly, and then somehow persuade them to stop hooking up so much with the ladies, you could get a WP48 advantage.

Also, I have a system for evaluating general managers. Each guy starts with 1 point. If they signed Jeff Green to a multi-year deal, however, then they end up with -50 points. That’s it. The elegance is in its simplicity.
Dave says:

December 21, 2012 at 3:24 pm

Mike G,
Now that is funny!

M	T	W	T	F	S	S
« Nov				Jan »
					1	2
3	4	5	6	7	8	9
10	11	12	13	14	15	16
17	18	19	20	21	22	23
24	25	26	27	28	29	30
31

David Berri Educates Us On John Hollinger

Share this:

Leave a Reply Cancel reply