Reply With Quote
I knew Hollinger was a good man all along.
Not that I put too much stake into it, but last year he had us low even though we were at 3, because many of our victories were by < 5 points. And sure enough in the PO we did not perform up to expectation. (manu was injured pretty much all of last season, so I dont count that as an excuse towards how badly we performed in the POs).
Hopefully this slow start will blow over.
http://espn.go.com/nba/hollinger/powerrankings
I knew Hollinger was a good man all along.
I was thinking about starting a thread regarding this subject. I know a lot of people make fun of Hollinger's "numbers" but usually he gets it right at the end.
come guys, if we decided to hate hollinger and his re ed number system, let not get all excited the spurs are high in his rankings
Indeed, that is it.
Bake (Albany, NY)
Are the Spurs finally getting some momentum with Pop tinkering with the line-up? (By the way is Keith Bogans really the right move to start even in a limited role?). Or are they just beating cupcakes and showing no ability to beat solid clubs. Are they on the market for any trade action?
John Hollinger (2:23 PM)
I think the Spurs are still evaluating. Right now they're in a stretch where they pay a bunch of bad teams so it's tough to tell exactly where they stand. I don't expect to see Bogans stay a starter for long, and I wouldn't rule out trade activity come February once they have a better idea of what they have.
http://espn.go.com/sportsnation/chat/_/id/30000
This is BipolarTalk.
I never decided to hate Hollinger, in fact I've always found interesting how he gets right predictions (based on his numbers) that seem ridiculous at the time he makes them.
And I think his PER system is the best (less crapy) stat to evalute a player's worth.
The Hollinger system has major flaws, even if it ends up putting the Spurs at #1.
The Spurs are horrible against decent teams, and has basically feasting off terrible teams this season. The defense and offense are extremely inconsistent, not only game to game, but within each game. Elite teams run their plays off instincts, making smooth plays every time, the Spurs are still thinking about the next move, creating lots of miscues.
Still not panicking, as it's only end of December. The Spurs have the potential to play really well by February, and the practices will need to be put to good use for that.
Spurs record against .500 or better teams: 2-10
Spurs Road Win/Home Loss +/- differential: Even
whos getting all excited, i just found it interesting that last year spurs were the 3 seed, but he had us pretty low. And this year, all we see is gloom and doom and he has us at 4. If you actually think anyone is raising champagne glasses over regular season performance you're deluding yourself.
Correction: Hollinger and his re ed number system has us at 4.
Hollinger is an idiot no matter what his numbers say. Case closed.
I don't see how you can get mad at Hollinger. He strictly uses formulas as a means to try and interpret the game. Sure, sometimes his numbers don't tell the whole story, but he never claims to try and use outside logic to define his findings. He just presents numbers. It is one side of the game and a very important one at that.
most of the people who are getting so defensive about hollingers numbers probably dont want to look like opportunistic toads because they scoffed at him in the past when the spurs were given a bad hand.
But really even now no one is saying spurs are the 4th best team in the NBA, all the post said was its interesting that the same hollinger who said we werent good when we were 3rd seed last year, is saying we are good when we are sucking this year.
+1 to DPG..
Hollinger never claims that his method is better than anything else..he even admits the flaws in his method..he just sticks to his formula and it's actually accurate a lot of the time..do you guys want him to just give his normal opinion like the rest of the horrible ESPN analysts that don't provide anything interesting to the discussion?..
the advanced stats have been kind to the Spurs so far..our efficiency differential is 6th in the NBA, right behind the Lakers..
obviously the problem has been our effort vs. playoff teams, but as I pointed out in the other thread, it's been due to our backcourt..I highly doubt our backcourt continues to play like they did in those games, they're way too talented for that to happen..
Exactly. The Spurs have problems, but the good news is that everyone knows what they are and there is a solid likelihood most of them have a good chance of being corrected.
TP, Dice and Jefferson can all correct their play more than likely to some degree. Manu, the other problem is the only real worrisome question mark at this point. TP is young and in his prime. Jefferson is still a great athlete and has a lot of ability. Dice is a notorious slow starter. Manu on the other hand is older and no one knows what he has left.
Also, some early problems have already started to correct themselves; Mason is picking up his play, Hill is starting to get a little better, Bogans is out of his shooting slump and Blair is getting solid minutes.
The Spurs bench is still extremely solid and should maintain, and one of the main questions, can Bonner be productive off the bench, was answered quite nicely.
Hollinger sounds crazy when he does not try to apply logic to his numbers, but the numbers do tell the story that the Spurs have the potential to be much better and they are still decent even with more than half of their stars struggling.
Well if you equate strictly winning to playing well, then it might seem odd. But when last year the Spurs were winning, there were a ton of close games they eeked out.
This year their point differential against the average teams is large and in their losses (for the most part) the point diff. is pretty close.
It is a means to try and guage the level, not just the win/loss total.
At least from what I recall, Hollinger did use his system to rank teams, he even write full articles to explain his numbers. The only exception I have seen are those Bruce Bowens of the world, where their defensive contribution do not directly lead to stats.
He kept harping on point differential as the most important sign of a championship, but he assigns these seemingly random weightings to each number. Strength of schedule, record in the last 10 games, things of that nature, I doubted that he ever went through the game by game progression of all the NBA teams of the past and see which variable are strong indicators of a championship team. He just came up with a bunch of formulas, either through observations or common sense, and then assign random weights.
That is not statistics at all.
He's right. Winning or losing close games don't tell you anything substantially different about the quality of a team. Last season the Lakers won 0.571 of their close games, only 11th league wide. Dallas and New Orleans were the leaders. And this isn't a statistical oddity.
Not that I put too much stake into it, but last year he had us low even though we were at 3, because many of our victories were by < 5 points. And sure enough in the PO we did not perform up to expectation. (manu was injured pretty much all of last season, so I dont count that as an excuse towards how badly we performed in the POs).
Hopefully this slow start will blow over.
http://espn.go.com/nba/hollinger/powerrankings
Last edited by mogrovejo; 12-29-2009 at 11:05 PM.
Yeah, it's the metric that better predicts play off success.
You mean he defined the linear weights arbitrarily? I doubt it. He didn't need to go game by game, he just needed to run regressions. That's why his formula doesn't account to stuff like injuries or possible matchups.but he assigns these seemingly random weightings to each number. Strength of schedule, record in the last 10 games, things of that nature, I doubted that he ever went through the game by game progression of all the NBA teams of the past and see which variable are strong indicators of a championship team. He just came up with a bunch of formulas, either through observations or common sense, and then assign random weights.
That is not statistics at all
yea i just checked his "formula" and the sucker has used 0.33 and 0.67. How easy is it to run a regression to assign weights based on the past? this guy is seriously overpaid to work for ESPN.
While point differential is (and I will take his word for it) a strong indicator of playoff success, this only applies to end of season results. The current records are moving targets, and different teams faced different sets of games under different conditions.
Hollinger is trying to predict playoff success with a moving set of data. The practice of taking regressions of an 82 game schedule and apply it to teams that has played 20, 30, 40 or 50 games is inaccurate, and just plain wrong. A team, after playing 82 games, are assured of playing a defined set of opponents.
However, naturally, the best team in the league will always face weaker compe ion (not playing against themselves), so other indicators such as strength of schedules are in.
Check this formula:
http://insider.espn.go.com/nba/insid...Rankings-Intro
Divide by 0.037 why? Multiply by 3.5 why? (I am guessing he wanted to move everything closer to the 100 area. Why is strength of schedule worth twice as much as strength of schedule in the last 10 games? Why are margins twice as valuable as margin in the last 10 games? Why are margins and strength of schedules given their respective weights?
I am not doubting the variables he used (although I am sure this is not the most accurate), I am doubting the weightings he used for each variable. It is OK if he would argue that this his automated (lazy) method of ranking teams, but don't masquerade it as some kind of scientific analysis.
I don't subscribe to all of Hollinger's methods, but I am just making the point that people hate on him for the wrong reason. If you want to hate on him, hate on him for his methods used to calculate, but not for the outcome and the numbers saying what they do.
There are currently 1 users browsing this thread. (0 members and 1 guests)
Posting Permissions
