Hollinger really gets on my nerves

[slamdunk]

Eh. I've learned not to pay attention to what these so-called experts think. If they really knew what was going to happen then they wouldn't be wasting their time working at ESPN.

[cordobes]

It's his opinion. Of course a team doesn't need the best record to be considered the better one. But why do you read him?

Well, it does not help when Cavs lose to the Celtics in their only meeting so far. I just don't think point differential proves too much at this point. I was not specifically looking to read Hollinger but I was reading the Daily dime and it was there.

What's historically the best predictor of future success at this point of the season, point differential or W/L record?

Well the only 2 teams with this record in the history of the game won the NBA championship so I guess it would W/L record of course.  Who had the best record at this point last year?

You didn't understand the question. Teams with Cleveland's point differential at this point never lost the championship either. It has been shown that Pythagorean expectation formula predicts future team winning percentage better than actual winning percentage.

It just means that the correlation is higher, from an historical perspective.

[cordobes]

You know I keep hearing that point differential is a greater predicter of success than win-loss record so what I did was go back and check the last 20 years to see if this was true.

Year        Champion       Runner Up       Best W-L      Best Pt Dif

1989        Detroit             Lakers              Detroit          Phoenix
1990        Detroit             Portland           Lakers          Phoenix
1991        Chicago          Lakers             Portland        Chicago
1992        Chicago          Portland          Chicago        Chicago
1993        Chicago          Phoenix           Phoenix         Seattle
1994        Houston          New York          Seattle         Seattle
1995        Houston          Orlando          SA Spurs       Seattle
1996        Chicago          Seattle            Chicago        Chicago
1997        Chicago          Utah               Chicago        Chicago
1998        Chicago          Utah               Chi/Utah        Lakers
1999        SA Spurs      New York         Utah/SA         SA Spurs
2000        Lakers           Indiana             Lakers          Lakers
2001        Lakers            76ers             SA Spurs       SA Spurs
2002        Lakers           New Jersey     Kings             Kings
2003        SA Spurs       New Jersey     SA/Dal           Dallas
2004        Detroit           Lakers               Indiana        SA Spurs
2005        SA Spurs      Detroit               Phoenix        SA Spurs
2006        Miami             Dallas             Detroit           DET/SA
2007        SA Spurs      Cleveland        Dallas           SA Spurs
2008        Boston           Lakers             Boston         Boston


So what does the last twenty years tell us about win-loss record versus point differential? Well:

- 9 times in the last 20 years the team with the best record became NBA champion
- 9 times in the last 20 years the team with the best point differential became NBA champion
- 11 times in the last 20 years the team with the best win-loss record made the Finals
-9 times in the last 20 years the team with the best point differential made the Finals
- 11 times the team with the best win-loss record had the best point differential
- only 6 times did the team that had the best win-loss record and best point differential win the NBA championship

So considering that the team with the best win-loss record has played in the Finals 2 more times than the team with the best point differential, that the team with the best win-loss records have won as many championships as the team with the best point differential and that the team with the best win-loss record had the best point differential more often than they actually won the championship, is it fair to say that point differential and win-loss record are a lot more evenly measured as a factor for success than Hollinger's Power Ranking formula says it is?

I would think so considering these facts.

http://armchairgm.wikia.com/NBA_Point_Differential_-_The_Most_Power_Stat

http://files-upload.com/361797/NBAPointsvs.Wins.xls.html

[Birdbrain]

It's his opinion. Of course a team doesn't need the best record to be considered the better one. But why do you read him?

Well, it does not help when Cavs lose to the Celtics in their only meeting so far. I just don't think point differential proves too much at this point. I was not specifically looking to read Hollinger but I was reading the Daily dime and it was there.

What's historically the best predictor of future success at this point of the season, point differential or W/L record?

Well the only 2 teams with this record in the history of the game won the NBA championship so I guess it would W/L record of course.  Who had the best record at this point last year?

You didn't understand the question. Teams with Cleveland's point differential at this point never lost the championship either. It has been shown that Pythagorean expectation formula predicts future team winning percentage better than actual winning percentage.

It just means that the correlation is higher, from an historical perspective.

No I understood the question I just countered it with a more valid point IMO.

[cordobes]

It's his opinion. Of course a team doesn't need the best record to be considered the better one. But why do you read him?

Well, it does not help when Cavs lose to the Celtics in their only meeting so far. I just don't think point differential proves too much at this point. I was not specifically looking to read Hollinger but I was reading the Daily dime and it was there.

What's historically the best predictor of future success at this point of the season, point differential or W/L record?

Well the only 2 teams with this record in the history of the game won the NBA championship so I guess it would W/L record of course.  Who had the best record at this point last year?

You didn't understand the question. Teams with Cleveland's point differential at this point never lost the championship either. It has been shown that Pythagorean expectation formula predicts future team winning percentage better than actual winning percentage.

It just means that the correlation is higher, from an historical perspective.

No I understood the question I just countered it with a more valid point IMO.

Ok, you didn't answer the question then.

[NoraG1]

Another reason I don't  buy Hollinger and his formulas and that the Cavs are playing better. Do the Cavs take a teams best shot because they are defending champs? No. The degree of difficulty is slightly higher for the Celtics imo.

[nickagneta]

You know I keep hearing that point differential is a greater predicter of success than win-loss record so what I did was go back and check the last 20 years to see if this was true.

Year        Champion       Runner Up       Best W-L      Best Pt Dif

1989        Detroit             Lakers              Detroit          Phoenix
1990        Detroit             Portland           Lakers          Phoenix
1991        Chicago          Lakers             Portland        Chicago
1992        Chicago          Portland          Chicago        Chicago
1993        Chicago          Phoenix           Phoenix         Seattle
1994        Houston          New York          Seattle         Seattle
1995        Houston          Orlando          SA Spurs       Seattle
1996        Chicago          Seattle            Chicago        Chicago
1997        Chicago          Utah               Chicago        Chicago
1998        Chicago          Utah               Chi/Utah        Lakers
1999        SA Spurs      New York         Utah/SA         SA Spurs
2000        Lakers           Indiana             Lakers          Lakers
2001        Lakers            76ers             SA Spurs       SA Spurs
2002        Lakers           New Jersey     Kings             Kings
2003        SA Spurs       New Jersey     SA/Dal           Dallas
2004        Detroit           Lakers               Indiana        SA Spurs
2005        SA Spurs      Detroit               Phoenix        SA Spurs
2006        Miami             Dallas             Detroit           DET/SA
2007        SA Spurs      Cleveland        Dallas           SA Spurs
2008        Boston           Lakers             Boston         Boston


So what does the last twenty years tell us about win-loss record versus point differential? Well:

- 9 times in the last 20 years the team with the best record became NBA champion
- 9 times in the last 20 years the team with the best point differential became NBA champion
- 11 times in the last 20 years the team with the best win-loss record made the Finals
-9 times in the last 20 years the team with the best point differential made the Finals
- 11 times the team with the best win-loss record had the best point differential
- only 6 times did the team that had the best win-loss record and best point differential win the NBA championship

So considering that the team with the best win-loss record has played in the Finals 2 more times than the team with the best point differential, that the team with the best win-loss records have won as many championships as the team with the best point differential and that the team with the best win-loss record had the best point differential more often than they actually won the championship, is it fair to say that point differential and win-loss record are a lot more evenly measured as a factor for success than Hollinger's Power Ranking formula says it is?

I would think so considering these facts.

http://armchairgm.wikia.com/NBA_Point_Differential_-_The_Most_Power_Stat

http://files-upload.com/361797/NBAPointsvs.Wins.xls.html

Except that the formulas given in the examples you gave were for point differential between offensive and defensive efficiency and not total point differential. Also, I want to see how much more point differential is a predicter of future success as compared to win-loss. From my very, very simple stats it is pretty even. But in Hollinger's numbers regular point differential(not point differential in efficiencies) far outweighs any other factor in his formula.

I need more proof than what your flawed examples provide.

[NoraG1]

We're 26-2, but in case you missed it, the Cavs are just 1 loss behind us. I understand homerism and all, but it's simply foolish not to be worried about the Cavs at this point. The difference between the two teams is, in all aspects, pretty small.

what happens when all 3 of these guys are playing 48 minutes per game?

As far as I am concerned, they probably get too winded to close the game, that's what happens.

I don't think it is homerism, I  would think saying a team on an 18 game streak and only 2 losses so far can't possibly be the best team in the league is questionable also even if I was not a Celtics fan.

[indeedproceed]

Anyone else really, really tired of this?

Anyone else start reading a point differential defense then immediately plug their fingers into their ears and start going

"la la la la I don't care until January 9th la la la I don't care until the Celtics play the cavs again la la la"

Cuz I have started doing that.


EDIT: Nick, good job though going out and getting the numbers.

[celticmaestro]

We should have learned by now to ignore these "experts".

[FanInTheSouth]

Foun this in the ESPN Daily Dime:

Tacoma, WA): Do you really think that the Cavs are a better team than the Celtics?

John Hollinger: I think the Cavs have played better than the Celtics. Whether they're a better team or just had a really hot 25 games is still to be determined. However, if I'm in Boston they're the one team that I'm deathly afraid of -- if they're this good with LBJ playing 35 minutes, what will they be like when he plays 48?


Getting really annoyed by Hollinger. How can he possibly think Cavs are better? Cavs have 2 more losses, lost to the Celtics and the Big 3 for the most part are not playing 35 min. each and are on an 18 game winning streak tying the longest in NBA history.  The Hollinger formula is not the end all, be all of determining how good a team is.

I think you mean tying the longest streak in Celtics history. 

Also, I try not to put too much value in what the experts say.  Remember that only 1 person picked the Celts to win the finals last year on ESPN.com, and we all know how that turned out.  It's easy for people to be anamored with the way the Cavs are playing, especially since they have Lebron.  This will all be settled out next May, until then I say let everyone drool over the Lebronaliers, maybe we can hit the playoffs as underdogs 

[orrzor]

I think Hollinger is right in that the Celtics should be worried about the Cavs. They were our toughest competition last season and they will likely be so again. And a win in a season-opener doesn't mean much in my eyes. But the Cavs should also be scared of us.

Also I think teams play harder against the defending champs than they do the King. I mean if you just look at writeups on other non-Celtics games, several players and coaches will continually say after a win they had a good game, but need to be playing as well as the Celtics. Celtics are the benchmark comparison and I think teams do try harder to beat us. Especially if you look at Atlanta, the Lakers, and the Cavs, they repeatedly say they need to defend as good as the C's.

[guava_wrench]

Why do people get bothered by the Hollinger rankings? He never said they were authoritative. Here is a quote from Hollinger:

Since this is an entirely automated ranking, you'll notice certain "human" factors missing.

It doesn't know which players are about to come back from injury or which teams have been playing without their best players for the past 10 games.

Along the same lines, it doesn't take into account injuries, trades, controversial calls or any other variables -- just the scores, please.

Nonetheless, it can be very useful because it allows us to see what the landscape looks like when we remove our usual filters.

It is meant to be useful, not authoritative. It is meant to be useful by NOT relying simply on W-L record, especially midseason when teams have played more or less home games or bad teams.

Considering how many more losses we had on the road last year, we can expect that we will have more losses when we play more games on the road, and our advantage in home games to start the season may have helped us to this record. These are the kinds of insights we can get from Hollinger's rankings.

Try asking yourself why his rankings would have us lower than the Cavs instead of feeling cheated.

Keep in mind that if we do not count the head-to-head game which we played in OUR arena when their team was not yet in sync, and if we discount the games when big Z was injured (esp considering we have been healthy all season), they also only have 2 losses.

[indeedproceed]

I think Hollinger is right in that the Celtics should be worried about the Cavs. They were our toughest competition last season and they will likely be so again. And a win in a season-opener doesn't mean much in my eyes. But the Cavs should also be scared of us.

Also I think teams play harder against the defending champs than they do the King. I mean if you just look at writeups on other non-Celtics games, several players and coaches will continually say after a win they had a good game, but need to be playing as well as the Celtics. Celtics are the benchmark comparison and I think teams do try harder to beat us. Especially if you look at Atlanta, the Lakers, and the Cavs, they repeatedly say they need to defend as good as the C's.

Second

[BballTim]

  An interesting stat is that the Celts are 14-1 against teams with winning records while the Cavs are 9-4.