CELTICS #1 In ESPN'S New Power Rankings - Both Stein AND Hollinger!

[nba is the worst]

Hard to believe, but true:
http://espn.go.com/nba/hollinger/powerrankings
http://espn.go.com/nba/powerrankings

[GreenFaith1819]

I'm shocked, especially with Hollinger's.

The funny thing is, with Hollinger's rankings we are only .094 pts in front of MIA. I bet that with his formula, if Paul had not gotten that triple double, we'd still be behind MIA .

[kozlodoev]

I'm shocked, especially with Hollinger's.

The funny thing is, with Hollinger's rankings we are only .094 pts in front of MIA. I bet that with his formula, if Paul had not gotten that triple double, we'd still be behind MIA .

We've been on top of Hollinger's power rankings for a while now. I don't see what the problem with Miami is, they've been on a tear lately.

[GreenEnvy]

Is this really surprising?

They have obviously been playing better than anybody right now and for the whole season.

[byennie]

I don't mind the Hollinger rankings, since at least you can look up the formula and read his rationale (it would be quite annoying if it has hidden).

I just wish they would take the W-L record off the page (since it's not part of the formula) and show how much each component contributes to the score.

For example, I ran this real quick on the Celtics (this is slight off, probably because I am using all of the rounded numbers from the page instead of the original data):

100+
(((SOS-0.5)/0.037)*0.67) = 0.25225
+ (((SOSL10-0.5)/0.037)*0.33) = -.005405
+ (0.67*(MARG+(((ROAD-HOME)*3.5)/(GAMES))) = 6.3867
+ (0.33*(MARGL10+(((ROAD10-HOME10)*3.5)/(10))))) = 4.433

Or in readable terms:

0.25 points for strength of schedule
-0.005 points for recent strength of schedule
6.39 points for scoring margin (weighted for home / away)
4.43 points for recent scoring margin (weighted for home / away)

If you saw that on the page, at least you'd know what they are getting "credit" for...

[GreenFaith1819]

I'm shocked, especially with Hollinger's.

The funny thing is, with Hollinger's rankings we are only .094 pts in front of MIA. I bet that with his formula, if Paul had not gotten that triple double, we'd still be behind MIA .

We've been on top of Hollinger's power rankings for a while now. I don't see what the problem with Miami is, they've been on a tear lately.

I think last week we were 2nd, behind MIA.

There was a similar thread discussing this last week, I think. My thing was - how could we be behind MIA (Last week) when we'd beaten them twice? And our SOS (Strength of Schedule, I think) was higher than theirs?

I haven't looked at all of his metrics, but I suppose that he has a methodology in place.

[guava_wrench]

I'm shocked, especially with Hollinger's.

The funny thing is, with Hollinger's rankings we are only .094 pts in front of MIA. I bet that with his formula, if Paul had not gotten that triple double, we'd still be behind MIA .

We've been on top of Hollinger's power rankings for a while now. I don't see what the problem with Miami is, they've been on a tear lately.

I think last week we were 2nd, behind MIA.

There was a similar thread discussing this last week, I think. My thing was - how could we be behind MIA (Last week) when we'd beaten them twice? And our SOS (Strength of Schedule, I think) was higher than theirs?

I haven't looked at all of his metrics, but I suppose that he has a methodology in place.

Hollinger's rankings change daily. We switched back and forth with them last week night to night. +1 against Was no doubt hurt them.

Head to head is irrelevant. Should Clev have been ahead of us when they were 1-0 against us? Of course not.

It would be nonsensical to consider head to head in a ranking like that since you will get situations where you have multiple teams that have a single direction cycle where team A is 1-0 against B, B is 1-0 against C, and C is 1-0 against A.

[indeedproceed]

I'm shocked, especially with Hollinger's.

The funny thing is, with Hollinger's rankings we are only .094 pts in front of MIA. I bet that with his formula, if Paul had not gotten that triple double, we'd still be behind MIA .

We've been on top of Hollinger's power rankings for a while now. I don't see what the problem with Miami is, they've been on a tear lately.

Hollinger Last week:
12/13/2010   1
12/14/2010   1
12/15/2010   1
12/16/2010   2
12/17/2010   1
12/18/2010   2
12/19/2010   1

Because we replaced the Knicks game (a 2pt win) with the Hawks game (23 point win) in our "Last 10 games" stats, our last 10 margin of victory went down enough to bring us even with Miami.

Check out the "power rankings breakdown" I do ever week (tues or wed on the front page) if you guys get a kick out of this stuff. This was last week's

http://www.celticsblog.com/2010/12/15/1877706/and-on-the-seventh-week-we-kicked-butt-i-said-its-power-rankings-time

[nba is the worst]

I'm shocked, especially with Hollinger's.

The funny thing is, with Hollinger's rankings we are only .094 pts in front of MIA. I bet that with his formula, if Paul had not gotten that triple double, we'd still be behind MIA .

We've been on top of Hollinger's power rankings for a while now. I don't see what the problem with Miami is, they've been on a tear lately.

I think last week we were 2nd, behind MIA.

There was a similar thread discussing this last week, I think. My thing was - how could we be behind MIA (Last week) when we'd beaten them twice? And our SOS (Strength of Schedule, I think) was higher than theirs?

I haven't looked at all of his metrics, but I suppose that he has a methodology in place.

Hollinger's rankings change daily. We switched back and forth with them last week night to night. +1 against Was no doubt hurt them.

Head to head is irrelevant. Should Clev have been ahead of us when they were 1-0 against us? Of course not.

It would be nonsensical to consider head to head in a ranking like that since you will get situations where you have multiple teams that have a single direction cycle where team A is 1-0 against B, B is 1-0 against C, and C is 1-0 against A.

I disagree. It would be simple to give extra weight in a formula to head to head results within the rankings, and your A -B-C scenario would be unchanged.

[nba is the worst]

I'm shocked, especially with Hollinger's.

The funny thing is, with Hollinger's rankings we are only .094 pts in front of MIA. I bet that with his formula, if Paul had not gotten that triple double, we'd still be behind MIA .

We've been on top of Hollinger's power rankings for a while now. I don't see what the problem with Miami is, they've been on a tear lately.

Hollinger Last week:
12/13/2010   1
12/14/2010   1
12/15/2010   1
12/16/2010   2
12/17/2010   1
12/18/2010   2
12/19/2010   1

Because we replaced the Knicks game (a 2pt win) with the Hawks game (23 point win) in our "Last 10 games" stats, our last 10 margin of victory went down enough to bring us even with Miami.

Check out the "power rankings breakdown" I do ever week (tues or wed on the front page) if you guys get a kick out of this stuff. This was last week's

http://www.celticsblog.com/2010/12/15/1877706/and-on-the-seventh-week-we-kicked-butt-i-said-its-power-rankings-time

I LOVE that weekly! 

[Fafnir]

I disagree. It would be simple to give extra weight in a formula to head to head results within the rankings, and your A -B-C scenario would be unchanged.

How would you do it?

I don't think it'd be simple at all, not with 30 teams in the league.

[nba is the worst]

I disagree. It would be simple to give extra weight in a formula to head to head results within the rankings, and your A -B-C scenario would be unchanged.

How would you do it?

I don't think it'd be simple at all, not with 30 teams in the league.

Not being a stats geek, I'm loath to create a formula, but it would be something along the lines of "head to head scoring margin (weighted for home / away)" added after the rest - automatically calculated like the others.

[green tea]

Great.  We know how accurate Stein and Hollinger are, so now I'm worried!  Hehe.

I'd rather fly below the radar for the whole regular season.  Let Miami win 15....20 in a row for that matter.  Let the Lakers win 58-62 games.  

If the C's are right in the mix with 55 wins and healthy (and no doubt ranked #5 or lower on these so called experts' lists), then I'm a happy man.

[Fafnir]

I disagree. It would be simple to give extra weight in a formula to head to head results within the rankings, and your A -B-C scenario would be unchanged.

How would you do it?

I don't think it'd be simple at all, not with 30 teams in the league.

Not being a stats geek, I'm loath to create a formula, but it would be something along the lines of "head to head scoring margin (weighted for home / away)" added after the rest - automatically calculated like the others.

That doesn't work.

The goal of the formula is to create a ranking for all 30 teams. So you'd need to consider head to head with each permutation of teams. That's 29 additional variables being added to the calculations. All of that added to the equation which wouldn't capture anything useful since your proposed idea isn't any different than just using the aggregate scoring margin.

Since you're adding all 29 possible combinations together (since you need to rank all 30 teams you're just going to get scoring margin)

[Change]

This is Great. We all know Power Rakings=Championships