Hollinger Power Rankings

John

has been known to love the new Bulls for a while now. This does not surprise me.

by Option27 on Jan 18, 2007 12:08 PM CST reply actions

Same here...

I was going to say, "The Bulls, unsurprisingly, are #5."

He takes a lot of stock in margin of victory. The Bulls' two 40-pt blowouts give them an additional 2 points of margin of victory right now. And they add up to one whole one for the entire season. That's huge. Although there is definitely something to be said for beating TWO teams by 40 points-as in, even when the Bulls puts its bench in when a game is in hand, it doesn't stop scoring.

by tyger1147 on Jan 18, 2007 12:25 PM CST up reply actions

exactly...

for anyone without insider or who doesn't know otherwise, Hollinger wrote an article a month ago maybe where he says basically that the Bulls are a lot better than their record indicates and the Cavs are worse and you can expect to see the teams diverge at some point accordingly.

Of course, since then, the Cavs have taken the division lead and held it for a while but I digress.

by tal1286 on Jan 18, 2007 12:37 PM CST up reply actions

Cavs

They've also lost 2 straight to Seattle and Portland.

by your friendly BullsBlogger on Jan 18, 2007 2:20 PM CST up reply actions

The rankings are based on a formula

by Paxson Jackson @ Blog a Bull on Jan 18, 2007 12:42 PM CST up reply actions

You missed the point

Hollinger, as far as I know, has no particular fan allegiance to the Bulls. He thinks average point differential, among other things of course, is a better indicator of a team than simple W/L record. Since he writes for ESPN based on his statistical analysis, I doubt he would have written, more than once, that he thinks the Bulls are better than what they get "credit" for if they weren't good at the stats he likes. It's not surprising he uses stats that he thinks are better indicators of a team than its W/L record. Thus, it's not surprising that the Bulls are ranked high in his power rankings.

by tyger1147 on Jan 18, 2007 1:58 PM CST up reply actions

How did I miss the point?

Was there a point to miss?

by Paxson Jackson @ Blog a Bull on Jan 18, 2007 2:05 PM CST up reply actions

Love Hollinger

Says here in the rankings that we have the 2nd best D in the league.

http://insider.espn.go.com/nba/hollinger/teamstats?sort=defeff&league=nba

by dcarioca on Jan 18, 2007 12:41 PM CST reply actions

That makes sence

The Bulls have quietly climbed back in the top 3 defensive field goal percentage category.

Their help defense is so unbelievable, I think us fans take it for granted sometimes. I guess you can say we're spoiled and we've come to expect it.

Other teams don't get to see that kind of defense often and when you hear other team's broadcasters watch the Bulls, they're just in awe.

by Option27 on Jan 18, 2007 12:47 PM CST up reply actions

it doesn't matter if hollinger loves the bulls

the rankings are based on a formula, and were hollinger trying to inflate the bulls, he would ahve had to do a ton of tinkering with the formula

by milesgmsu on Jan 18, 2007 1:18 PM CST reply actions

You missed it, too.

Hollinger has stats he likes, even before this formula.

He has written about how he likes the Bulls because they are good at the stats he likes.

He used the stats he likes in his power rankings.

Thus, it's not surprising the Bulls are ranked high in his power rankings.

Unless, he also has a special fan allegiance to the Bulls. But all I've ever read is that he likes the Bulls because they have a good average point differential, and beat the bad teams by a lot, rarely get blown out, etc.

by tyger1147 on Jan 18, 2007 2:02 PM CST up reply actions

I'm really confused now

by Paxson Jackson @ Blog a Bull on Jan 18, 2007 2:06 PM CST up reply actions

what he's saying

is that hollinger likes a certian set of stats, the bulls have these stats, thus hollinger likes the bulls

a=b b=c a=c

basically that the forumla was written for stats where the bulls excel at....while if the stat was

number of highlights shown on sportscenter and shoe deals....hollinger would like the lakers and heat and thus the heat and lakers would be high in hsi power rankings

by milesgmsu on Jan 18, 2007 2:35 PM CST up reply actions

I think I understood that part,

I'm really confused why someone would come to that conclusion.

by Paxson Jackson @ Blog a Bull on Jan 18, 2007 2:37 PM CST up reply actions

the conclusion is logical

and correct....perhaps we're arguing the same point and if we are i apologize in advance but:

holinger has a set of stats he likes, the bulls excell in those areas (as opposed to others), thus logically when hollinger would build a forumla that he beleives ranks the teams, and builds into the forumla the stats he loves and the bulls excel at, its only logical the bulls would place high

by milesgmsu on Jan 18, 2007 2:40 PM CST up reply actions

What?

Dude, I don't think quantifying strength of schedule has anything to do with a wild guess that Hollinger is a Bulls fan.

by Paxson Jackson @ Blog a Bull on Jan 18, 2007 2:43 PM CST up reply actions

i'm missing your point

and furthermore, i'm confused...are we arguing?

by milesgmsu on Jan 18, 2007 3:00 PM CST up reply actions

HAPPY

First Transitive Property Description on Blog-A-Bull!!!!!!!

Deng, he's good!!!!!!!!!

by cubbybear on Jan 18, 2007 8:23 PM CST up reply actions

no offense but...

That dosent make any sense. What stats does he "prefer" over others? In fact the whole gist of Hollinger's stats is that they are weighted in order to be compared fairly. The premise behind that is that you can't just look at pts, asts, t/o, etc.... since you have to take minutes in to account, among other things. Another example are assists. What's more important, the pass that leads to the basket, or the pass before that allowed the "assist" to happen? Therefore assists are weighted down for that, and usage rate. Overall it allows you to compare any player to any other player (same for teams).

by dcarioca on Jan 18, 2007 6:18 PM CST up reply actions

You're talking about player stats

Hollinger uses team stats for his power rankings. Hollinger makes the assumption that certain statistics (margin of victory) are better indicators of success than wins and losses. For many fans, wins and losses will always be the ultimate rhetorical trump card.

by Scotter on Jan 18, 2007 6:32 PM CST up reply actions

it's an assumption based on research

by your friendly BullsBlogger on Jan 18, 2007 6:51 PM CST up reply actions

No offense taken...

since you have no idea what you're talking about.

by tyger1147 on Jan 18, 2007 6:51 PM CST up reply actions

I'm still confused

I have no idea what you're trying to communicate here.

by Paxson Jackson @ Blog a Bull on Jan 18, 2007 6:53 PM CST up reply actions

Hey knucklehead!

This is the ranking explanation. Please stop telling people they missed a point about something, or whatever it is you know that we so obviously don't.

by Paxson Jackson @ Blog a Bull on Jan 18, 2007 7:01 PM CST up reply actions

allright.

Point taken. still, can anyone explain how these "preffered" stats favor the bulls?

by dcarioca on Jan 18, 2007 7:12 PM CST up reply actions

That's what I couldn't get around

I think 'tyger123456' or whatever has been confusing preference with reason.

by Paxson Jackson @ Blog a Bull on Jan 18, 2007 7:34 PM CST up reply actions

ahh i understand

hollinger likes margin of victory...this favors 06-07 bulls because we have 2 40 point victores, couple of 30 point victories, few 20, and a bunch of teens....the bulls have one of the best margin of victory in the league....furthermore, bulls have beaten a number of good teams latley, and as soon as the recent success rate changes to last 25 games, we are going to shoot up the rankings

by milesgmsu on Jan 18, 2007 8:55 PM CST up reply actions

Just to clarify

it's not last 25 games, it's last 25% of games once the Bulls have played at least 40 games. Depending on whether Hollinger rounds properly or just rounds up, that will be roughly the last 11 games up till the Bulls have played 44, then 12 games up to 48 games played, and so forth. So while recent play still helps Bulls, there won't be any shooting up the ratings. (I'm not sure you can really shoot up from # 5 anyway, can you?)

by wjb1492 on Jan 19, 2007 11:23 AM CST up reply actions

i misread it it

thats why i thought it was weird that he was doing 25 onceyou hit 40....percent makes much more sense

by milesgmsu on Jan 19, 2007 11:45 AM CST up reply actions

Who's on first?

No, he's an undersized power forward we should trade.

by bullshooter on Jan 19, 2007 1:42 PM CST up reply actions

The Ranking

has nothing to do with whether or not he likes the Bulls; it's a very specific formula. But in the comments for the Bulls he did say "My team...".

Fleece the Pig, Flog the Pony

by james@lifeinthecell on Jan 18, 2007 2:07 PM CST reply actions

That was some dude named Nathaniel,

or Nate Dogg as I like to call him. I'm really not sure what the comments are for, but it appears you can submit your own.

by Paxson Jackson @ Blog a Bull on Jan 18, 2007 2:09 PM CST up reply actions

Double checked...

I didn't notice that; just thought it was Hollinger all along. Thanks for the heads up.

Fleece the Pig, Flog the Pony

by james@lifeinthecell on Jan 18, 2007 2:11 PM CST up reply actions

Its updated daily

I assume Hollinger had no interest in coming up with daily comments, so instead they are letting the general public submit comments every day and choosing the "best" one.

B-jizzle my nizzle!

by preverbal on Jan 18, 2007 3:01 PM CST up reply actions

As far as I know...

Hollinger is a Hawks fan. Not that this has any impact on his rankings either...

"Hollinger makes the assumption that certain statistics (margin of victory) are better indicators of success than wins and losses. For many fans, wins and losses will always be the ultimate rhetorical trump card." - sbulls9030

For many fans, wins and losses are the trump card. What many fans don't understand is that margin of victory really is a better indicator of how good a team is. That's why he "makes that assumption" - because it just happens to be true.

by rangerforlife on Jan 18, 2007 7:29 PM CST reply actions

I have a problem

If wins and losses don't give us much info, then how does strength of schedule get factored in? Why not base the score on the opponents' scores, and let the loop work itself out. Anyway, why include it as even part of it.

Deng, he's good!!!!!!!!!

by cubbybear on Jan 18, 2007 8:26 PM CST reply actions

ahhhh...

now THAT's a good question.

by your friendly BullsBlogger on Jan 18, 2007 8:47 PM CST up reply actions

Just a guess...

He probably didn't include it because it matters a lot less in the pros than it would in college. Everyone plays everyone else at least twice, so aside from the difference between the two conferences, SOS probably comes out negligible. Plus, most people are already turned off by math and I doubt Hollinger wanted it any more complicated than it already was.

by rangerforlife on Jan 18, 2007 9:42 PM CST up reply actions

I disagree

Everyone might play everyone twice but they play their division foes double that ammount and there are big differences in some of the divisions, so 40 games in I would say strenght of schedule would have some effect. Especially with some home / away variable added in

by NY Chicago Fan on Jan 19, 2007 7:37 AM CST up reply actions

Wow... I didn't realize it was this complicated.

I didn't reply to Paxson Jackson because I just thought he was being the usual sarcastic, ridiculing jerk he usually is. He likes to swoop in and make deriding comments to others, while still typically displaying a decent knowledge of basketball statistics. I figured that's what he was doing here. So I ignored it.

But anyway, here's what I mean:
John Hollinger has either figured out for himself, or read someone else's work, that average scoring margin for a team is a lot more indicative of a team's future probable success than simple wins/losses. (This is somewhat common among basketball staticians. Matt sort of mentioned this earlier in the season saying something to the affect, "Some people don't think we should get excited about the Bulls blowing out these bad teams. But blowing out bad teams by 15 is better than beating them by five.")

So... Hollinger thinks scoring margin is important.

Hollinger has been lauding the Bulls all season. Since he attempts to avoid "the gut", this has probably been mostly due to the Bulls having had two huge wins, in terms of point differential, early in the season, Miami by 42 and Milwaukee by 25.

We following? Hollinger puts emphasis on scoring margin. Bulls have good scoring margin. Hollinger lauds Bulls.

Then, ESPN comes and asks Hollinger to do a Power Rankings system based purely on his stats, no "gut instinct" involved. Since Hollinger loves scoring margin, he makes it a big part of his statistical rankings, which also have heavy emphasis on "Last 10".

The Bulls have a 45 point victory w/in their last 10. This helps them a lot (otherwise, their last 10 would hurt them a lot with that last slide).

Therefore, if one is to expect a "stat guy" to make statistical power rankings, one would expect him to use one of his favorite stats, scoring margin.

Since the Bulls have a Top 4 scoring margin despite their "average" record, IT IS NOT SURPRISING they rank high on his power rankings.

To summarize:
--Hollinger thinks scoring margin is important
--Bulls have great scoring margin early in season
--Hollinger lauds/loves Bulls early in season
--Hollinger makes power rankings
--Hollinger includes scoring margin in power rankings
--Bulls STILL have great scoring margin
--Bulls rank high in Hollingers rankings

tyger11213294712884 is NOT SURPRISED.

Others, however, are baffled.

by tyger1147 on Jan 18, 2007 9:30 PM CST reply actions

A fair question, then, is...

...does one think the rankings accurately reflect how well teams are playing right now? Since I think the principle is correct, I do. But I see the whole thing as more of an amusement than anything else.

by rangerforlife on Jan 18, 2007 9:48 PM CST up reply actions

I'm still baffled how I missed the point

I never denied these are the Hollitron 5000 Power Rankings. You kinda ran wild on me for no reason. Scoring margin is a tried and true method in ranking/predictive systems for other sports, this isn't Hollinger hoping something sticks. You've been confusing so many people because words like "favorite" make Hollinger's rankings appear to be made up of arbitrarily selected stats. That's not the case.

by Paxson Jackson @ Blog a Bull on Jan 18, 2007 10:05 PM CST up reply actions

What?

Yeah, the brick-shooting backup point guard.

by bullshooter on Jan 19, 2007 1:44 PM CST up reply actions

Very Interesting

Thank you "Preverbal" - this is interesting. It is a reason I really enjoy this blog. I hope it is predictive of Bulls playoff success.

chgobr

by chgobr on Jan 19, 2007 7:31 PM CST reply actions

SoS

If he believes that margin of victory is a better predictor of a team's success than win%, shouldn't he use that in determining strength of schdeule too?

Great blog, btw. I've been a silent spectator so far, but enjoy the discussions.

by bullsfaninbigapple on Jan 19, 2007 8:57 PM CST reply actions

Strength of schedule

I agree - strength of schedule is critical. The Bulls are a good example. 3-9 when the schedule was very challenging. It seems this ranking does take that into account via the mechanisms below.

SOS = Season Win/Loss percentage of team's opponents, expressed as a decimal (e.g., .500)

SOSL10 = Season Win/Loss percentage of team's last 10 opponents, expressed as a decimal (e.g., .500)

HOME = Team's home games

HOMEL10 = Team's home games over the last 10 games

ROAD = Team's road games

ROADL10 = Team's road games over the last 10 games

GAMES = Team's total games

chgobr

by chgobr on Jan 20, 2007 9:24 AM CST up reply actions