The Official NBA Regular Season 07-08 Thread

[lastchance]

Don't worry GKA. Hollinger is the be-all end-all.

Also, I mean, yes, the Mavs and Suns were damn good, but like, the same reason the Spurs are the best team in the regular season last year is the same reason the Red Sox were the best team in the regular season.

PF-PA correlates with how good a team is, whether or not a team is beatable.

And over 82 games, it's a damn strong measure.

[Victor]

[ QUOTE ]
People really don't think the spurs were the best team in the league in the regular season last year?

I mean, come on ppl, this is 2p2, you guys should know about point differential by now.

[/ QUOTE ]

i dont understand why they even keep track of wins/losses

[Green Kool Aid]

[ QUOTE ]
Christian (New York): John, If the Spurs are prone to losing closer games, how is that a good thing? Isn't that the test of a tough, proven team? Sure the Bulls and Raps can blow out teams in the Leastern Conf. but how does that make them better than the Jazz who play tougher teams and win close games all the time.

SportsNation John Hollinger: Have had a lot of questions along these lines after my blog item on the power rankings. It all depends on whether you think the Spurs and Mavs performance in close games is predictive of how they'll do in future close games, or if it's just a fluke. I say fluke and I've got reams of data to back me up on this.

[/ QUOTE ]

[kidcolin]

lastchance,

PF-PA? Enlighten me.

GKA,

re: Hollinger. Ultimately he's awesome. But he has his flaws. His stats aren't impervious. I know less about his power rankings, but take PER for instance. It obviously doesn't account for defense very well. And it overvalues scoring. It balances the scoring bias somewhat, but players can still increase their PER simply by chucking, even if it's at the detriment to the team. For instance, Marbury fares very well under the PER metric, with a career 19 mark. Jason Kidd has a lower career mark by just a few decimals, even though Marbury has been pretty bad the past 3 seasons. Obviously, in terms of real value, Kidd blows Marbury out of the water.

[lastchance]

PF-PA is point differential. Points for and points against. See how advanced and complicated my awesome stat for determining how good a team is?

I'm not going deep into the Hollinger bag of tricks here.

Victor, Wins/losses do matter. Because that's how we determine the winners. That's not what we're talking about here.

[kidcolin]

oh.. I just never saw it labeled like that. I agree that it's a very strong measure taken over 82 games.

[tuq]

Are a lot of you posters Jerry's Kids? It appears as if so.

- Spurs may or may have not led objective NBA measures last year. This doesn't change the fact that

- they gave up home court advantage to both the Suns and Mavericks in their respective rounds, which unlike other sports actually matters in the NBA

- there is no predictive measure for 1 - the Mavs getting bounced in a perfect storm against the Warriors, 2 - Nash getting head-butted in game 1 and sitting out most of the rest while Barbosa (LOL) tried to lead the offense, and 3 - most importantly, two of their key stalwarts being banned for one of the most inexplicable offenses ever. You think it was legitmate? Fine. But stop factoring it into the series outcome like it was expected.

The Spurs got really lucky. Like poker, luck is a component of winning. Thus applying absolutes is ridiculous.

[lastchance]

1. Yes, because they were the best team in the NBA in the regular season.

2. WTF does this have to do with who is the best team?

3. WTF does this have to do with who is the best team?

[Artdogg]

Stop acting like PD is the end of all means, some teams handle garbage time differently than others, which can have a drastic effect on overall PD. Also there's the injury factor that tuq mentioned. Hollinger likes to assume it all evens out, which isn't true, which is why he was wrong about the best teams last year.

[tuq]

lastchance,

Statistics and logic answer #2 and #3? [img]/images/graemlins/confused.gif[/img]