John Hollinger

[dc]

-= original quote snipped =-


whats wrong with that? thats pretty much the same thing i (as well as many other analysts and even posters here) said at the time.

Nothing was wrong with the statement itself, it's just that he made that comment based all on the PER of all players involved (again, that was just a sample. The entire article contained his PER number crunching).

[Doctor MJ]

My take as a stat guy:

There's nothing wrong with PER, it's a useful stat, but Hollinger loses all credibility with the way he tries to make it more important than it is.

[lukeridenour]

its not a terrible statistic. i mean it isnt better than watching tapes of the player play but sometimes theres no time for that. perhaps per isnt 100% accurate but its not a bad measuring tool.

i like he durant and young exampe because it does show the flaws of the system. im sure of young was in durants position his per would not be as nearly as good. however, it does tell us that young is a fine contributer and that durant needs some major help and or improvement in his offensive efficiency.

so i dont we should believe in the per 100% but its a nice stat to look at if you arent taking it too seriously.

[nate33]

You guys don't get what Hollinger is trying to do.

Anybody can sit there are regurgitate conventional wisdom. Hollinger goes to the other extreme. He uses stats almost exclusively in an effort to avoid the biases of the herd. It's his niche. There's nothing wrong with that. I'm sure he is well aware of the limitations of PER but he clings to the numbers to inspire debate.

[HarlemHeat37]

My take as a stat guy:

There's nothing wrong with PER, it's a useful stat, but Hollinger loses all credibility with the way he tries to make it more important than it is.

agreed, but that's what most people would do..he's just hyping up his own creation/product..

[Doctor MJ]

-= original quote snipped =-



agreed, but that's what most people would do..he's just hyping up his own creation/product..

The other big names in the basketball stat biz are actually much, much better about it.

[Doctor MJ]

You guys don't get what Hollinger is trying to do.

Anybody can sit there are regurgitate conventional wisdom. Hollinger goes to the other extreme. He uses stats almost exclusively in an effort to avoid the biases of the herd. It's his niche. There's nothing wrong with that. I'm sure he is well aware of the limitations of PER but he clings to the numbers to inspire debate.

I can't let him off the hook so easily because he doesn't just limit discussion to stats, he largely limits discussion to *his* stats. I think you've laid out a great defense for stat guys in general, but Hollinger should not be given the same immunity.

[JGeils]

Hollinger writes some interesting articles, etc., but the problem is that when his formulas tell him something that makes no sense he assumes that the formula is right and he has just discovered something else that everyone is missing. It never seems to occur to him that maybe his formulas don't reflect who the best teams or players in the league are.

If his power rankings told him the Grizzlies are the best team in the league he wouldn't re-evaluate his system, he would write a column about how the Grizzlies are the best team in the league and no one realizes it but him.

If he was more willing to adjust when his math doesn't reflect say anything meaningful about reality people would hate on him less.

Not true in the slightest. And anyone with a cursory knowledge of his books, scouting reports, or columns would know better.

Statements like these usually come from people who get a good strawman idea in their head, they don't own Insider, and they just sort of run with it because it sounds good.

I can safely say that I watch and re-watch more games than any of the mainstream NBA scribes out there. But I can also safely say that Hollinger watches and re-watches (and, through the use of Synergy Sports, scouts) more NBA games that each of his ESPN NBA colleagues ... combined.

The guy works his ass off for you. And if his numbers don't make sense to him, he questions it. Has since 2002.

[JGeils]

Also, have any of you scouted Diogu?

It's not Hollinger's fault that Diogu:

1. Put on a ton of weight this summer.
2. Didn't get the minutes he deserved last year.

I have an account at Synergy Sports, which takes game tape of each individual player and allows you to pull up every single offensive possession (of specific possessions, like, "right block," "screen/roll," "left block") of a player. Going through Diogu's possessions last year: he was great. A hell of a player. Deserved way more minutes.

This year, not so much. Fatty.

[G35]

You guys don't get what Hollinger is trying to do.

Anybody can sit there are regurgitate conventional wisdom. Hollinger goes to the other extreme. He uses stats almost exclusively in an effort to avoid the biases of the herd. It's his niche. There's nothing wrong with that. I'm sure he is well aware of the limitations of PER but he clings to the numbers to inspire debate.

-= original quote snipped =-



I agree. The main reason some people don't like PER is because their favorite player's PER does not compare well to their favorite player's "rivals."

If it was just to inspire debate and show another POV then PER/Hollinger would be under less criticism. It's just that somehow, someway Hollinger got PER to become the new standard on how we rank players. Similar to the QB rating doesn't alwasy give the best representation of a players ability/impact.

People now just point to PER and now we have actually taken debate out of comparisons. Many people just go to PER and get their answer.

Which makes me curious to peoples reactions to PER when we compare say David Robinson (career PER 26.18 4th all time) to say Hakeem Olajuwon (career PER 23.59 19th all time).........

[loserX]

You can tell more from looking at stats than watching the games. Anyone who disagrees has a seriously overinflated ego. Your mind can't even come close to comprehending everything that stats can tell you. Not to mention, stats measure every second of a players gametime, something that's impossible for a fan to do.

And that's why nobody ever hires scouts.

Look, stats are interesting. They are also useful. But they are also approximations of reality...nobody (including Hollinger, as I hope he'd admit) has come up with a perfect statistical representation of player performance, it's simply too complex. Go ahead, use stats...but don't rely on them 100%. Integrate them with other techniques, like watching games.

[Ayt]

Read his scouting reports. Only an idiot would think Hollinger doesn't watch a ton of game.

[JGeils]

-= original quote snipped =-



And that's why nobody ever hires scouts.

Look, stats are interesting. They are also useful. But they are also approximations of reality...nobody (including Hollinger, as I hope he'd admit) has come up with a perfect statistical representation of player performance, it's simply too complex. Go ahead, use stats...but don't rely on them 100%. Integrate them with other techniques, like watching games.

And anyone who would actually bother to read Hollinger's work would know that Hollinger hasn't come anywhere close (in a decade of having his work appear online, in print, or both) to claiming that people should "rely on them 100%," or even half that.

Again, it's a moronic strawman argument that that doesn't actually hold serve in reality. Only put forth by those too lazy to research who they're talking about.

[Duiz]

He loves Chrissy Paul over Derawn Villiams.