All NBA teams announced

[hova-]

The NBA has announced the annual All NBA teams : http://www.nba.com/2013/news/05/23/lebr ... index.html

1st team:

G Chris Paul
G Kobe Bryant
F LeBron James
F Kevin Durant
C Tim Duncan

2nd team:

G Tony Parker
G Russell Westbrook
F Carmelo Anthony
F Blake Griffin
C Marc Gasol

3rd team

G James Harden
G Dwyane Wade
F Paul George
F David Lee
C Dwight Howard

Love to see TD in the first team. He is still so valuable. Blake should rather be in the third team imho, but with Love and Dirk out, he sneaked into 2nd team. I also think Tony Parker should be considered for the 1st team, but Paul and Kobe had once again great individual seasons.

Growing on Paul George after his clutch outing against Miami the other night. Reminds a bit of Scottie Pippen, my fav player of all time. Good to see him in the third team. I don't like Lee, but I guess he has helped the Warriors a lot during their run this season.

Biggest snubs for me: Joakim Noah, Zach Randolph

[Leftos]

NBA Stats Tracker's All-NBA Teams
- Season performances
- At least 60% of team's games played
- At least 10MPG
- Based on GmSc metric and position
- Stats and position as on Basketball-Reference.com

[NovU]

I'm not happy with Kobe inclusion in the 1st team. Harden in all regard had a better season and Wade arguably had a better season as well. He should be more of a 3rd team. The heck, Russell was better than Kobe.

I like Tim Duncan getting recognition. We probably don't have much time left to enjoy him playing at such high level with him nearing 40 pretty soon. The best PF ever to play the game. Gotta appreciate him even more this season.

Biggest snubs for me: Joakim Noah, Zach Randolph

Zach was terrible during the season though. He had issues with coach as well, thus starting from bench at times. Marc was the man for the Grizzlies. As for Noah, I loved him. I thought he was a DPOY otherwise for injury at the end.

[SoF'nAwesome]

Duh! How did I miss this?

viewtopic.php?f=21&t=91562

[Andrew]

Solid picks for the most part. It's a tough pick between the guards; Harden could probably at least be on the second team, but it's hard to drop Parker or Westbrook back to the third team.

[Spree#8]

Definitely better than the All-D teams. Harden should be swapped with Kobe though.

When was the last time a 45-37 team (or a 7th seed) had two All-NBA players?

[Andrew]

When was the last time a 45-37 team (or a 7th seed) had two All-NBA players?

The Knicks, last year.

[hova-]

So I looked up Tony Parker's vita just because the Spurs are really doing well lately and I have to say that I'm shocked that this is only his third appearance in an All NBA team. (2x 2nd team, 1x 3rd team)

This guy is so underrated it hurts. A 6-2 guard shooting 52% from the field this season (20ppg) and 49% for his career ... dunno how often you have those numbers for a PG (apart from some pass-first guards like Nash, Stockton or even Calderon).

Don't know about his defense exactly but I guess even if it is only mediocre I'd still take Parker over Westbrook all day long (not considering the futur with their given age).

Is it the Euro-thing? Is it the fact that nobody likes the Spurs? Why is there no love for TP?

[benji]

No love? The guy got on the second team this year.

It's simply how things work. There's Paul, Bryant, Wade who we all admit are instantly superior most seasons. That's three slots on two teams. Then you have Deron, Nash, Billups, Harden, Westbrook who have all put up years here and there at or near whatever Parker has done every year.

Parker hasn't exactly tore it up every single year. He has maybe four "legit" discussion years in his career. With this being his probably best year in all. (He's not better than Paul, ever, and so then most voters probably looked for a SG. And Wade/Kobe have topped him most years.)

Worth noting that like his fellow scorer benchmark Kevin Johnson, Parker doesn't shoot threes well, so he doesn't take them. Though Parker, along with teammate Manu, are for the era elite scoring guards.

And of course there's a Spurs fudge to it, Duncan has never won a DPOY. That's probably only four or five less than he deserves.

- Based on GmSc metric and position

[benji]

Cube root of PER*WS*WS/48:

First: Paul, Harden, LeBron, Durant, Marc Gasol
Second: Westbrook, Wade, Griffin, Anthony, Lopez
Third: Parker, Bryant, Ibaka, West, Duncan

Double counting PER:
First: Paul, Harden, LeBron, Durant, Lopez
Second: Westbrook, Wade, Griffin, Anthony, Duncan
Third: Parker, Bryant, Ibaka, West, Marc Gasol

Double counting WS/48:
First: Paul, Harden, LeBron, Durant, Marc Gasol
Second: Westbrook, Parker, Griffin, Anthony, Lopez
Third: Wade, Stephen Curry, Ibaka, Splitter, Chandler

[_Steve_]

Little surprise to see Joel Anthony on the 2nd team.
But Brook Lopez really surprises my. Hadn't had him on my radar.

[Leftos]

Cube root of PER*WS*WS/48:

First: Paul, Harden, LeBron, Durant, Marc Gasol
Second: Westbrook, Wade, Griffin, Anthony, Lopez
Third: Parker, Bryant, Ibaka, West, Duncan

Double counting PER:
First: Paul, Harden, LeBron, Durant, Lopez
Second: Westbrook, Wade, Griffin, Anthony, Duncan
Third: Parker, Bryant, Ibaka, West, Marc Gasol

Double counting WS/48:
First: Paul, Harden, LeBron, Durant, Marc Gasol
Second: Westbrook, Parker, Griffin, Anthony, Lopez
Third: Wade, Stephen Curry, Ibaka, Splitter, Chandler

You know WS remains a mystery to calculate. If I could implement it, I would've done so a long time ago. PER could work with MPG and Games Played lower limits, I guess. Didn't think of it. I don't see THAT many differences in the teams you've produced to what GmSc produces though, to be honest. I'm not saying that's an excuse to use such a simple metric as GmSc. Reason I started by using GmSc is because NBA 2K doesn't keep Opponent stats, so they have to be calculated from box scores. No opponent stats means no PER. But since B-R.com offers them, I'll add the option to calculate the best teams based on PER, if the user wants to.

Let me know if you ever decode WS.

[benji]

Wasn't directed at you, just made me actually see what it looked like this year.

[Leftos]

Wasn't directed at you, just made me actually see what it looked like this year.

What about the frowny face as far as the "GmSc and position" line of mine that you quoted?

[hova-]

No love? The guy got on the second team this year.

It's simply how things work. There's Paul, Bryant, Wade who we all admit are instantly superior most seasons. That's three slots on two teams. Then you have Deron, Nash, Billups, Harden, Westbrook who have all put up years here and there at or near whatever Parker has done every year.

Parker hasn't exactly tore it up every single year. He has maybe four "legit" discussion years in his career. With this being his probably best year in all. (He's not better than Paul, ever, and so then most voters probably looked for a SG. And Wade/Kobe have topped him most years.)

Worth noting that like his fellow scorer benchmark Kevin Johnson, Parker doesn't shoot threes well, so he doesn't take them. Though Parker, along with teammate Manu, are for the era elite scoring guards.

And of course there's a Spurs fudge to it, Duncan has never won a DPOY. That's probably only four or five less than he deserves.

- Based on GmSc metric and position

Well with no love I'm rather referring to the past than to the presence as you are right, a 2nd team selection is definitely showing some value. So you definitely answered my questions. The Spurs fudge + him not tearing it up every single year.

It would just be fun to see how the Spurs would have gone with any other PG than Parker. Like the ones you mentioned: Nash, Billups, Deron ... too bad it's not possible to find out. Because somehow I think that Parker is exactly the right player for the Spurs and maybe also the other way around.

If Parker had played for the Bucks his entire career, he maybe would have written "Monta Ellis" all over him (or name any other combo guard who has no success but decent numbers).

[Leftos]

Well, here I go again.

NBA Stats Tracker's All-NBA Teams
- Season performances
- At least 70% of team's games played
- At least 25MPG
- Based on PER metric and position
- Stats and position as on Basketball-Reference.com







Hint: Lower the minutes per game requirement, and this happens:



Fucking A.

But yeah, still doesn't compare to having Win Shares and doing the math magic of multiplying it with Per 48 and PER. Don't know why you'd cube-root it, since you don't actually care about the resulting number, but just the comparison of it between players.

[Patr1ck]

means neutral.

Also, fuck win shares. Just looking at the equations tells me it's convoluted.

[NovU]

Cube root of PER*WS*WS/48:

First: Paul, Harden, LeBron, Durant, Marc Gasol
Second: Westbrook, Wade, Griffin, Anthony, Lopez
Third: Parker, Bryant, Ibaka, West, Duncan

Double counting PER:
First: Paul, Harden, LeBron, Durant, Lopez
Second: Westbrook, Wade, Griffin, Anthony, Duncan
Third: Parker, Bryant, Ibaka, West, Marc Gasol

Double counting WS/48:
First: Paul, Harden, LeBron, Durant, Marc Gasol
Second: Westbrook, Parker, Griffin, Anthony, Lopez
Third: Wade, Stephen Curry, Ibaka, Splitter, Chandler

Looks great. Shows effectiveness of PER and winShare. Though it's not perfect, you can't entirely ignore its effectiveness, especially over 'eye test'. Now if somebody could come up with a metric that measure true essense of defensive impact.

[benji]

I suppose you could include DWS, that gives you:
First: Paul, Westbrook, LeBron, Durant, Marc Gasol
Second: Harden, Wade, Griffin, West, Duncan
Third: Conley, Curry, George, Anthony, Lopez

Double counting PER, WS/48 and including DWS and DRtg:
First: Paul, Westbrook, LeBron, Durant, Marc Gasol
Second: Harden, Wade, Griffin, West, Duncan
Third: Parker, Curry, George, Anthony, Lopez

What about the frowny face as far as the "GmSc and position" line of mine that you quoted?

Like Pdub said it's the neutral face and I'm not sure how well GameScore scales to a seasonal level.

Well, here I go again.

Fucking A.

But yeah, still doesn't compare to having Win Shares and doing the math magic of multiplying it with Per 48 and PER.

Try some function of Minutes and PER (like Minutes * PER or % of Possible Minutes Played * PER) and see what you get.

Don't know why you'd cube-root it, since you don't actually care about the resulting number, but just the comparison of it between players.

So they're equally weighted.

[shadowgrin]

Cube root of PER*WS*WS/48:

First: Paul, Harden, LeBron, Durant, Marc Gasol
Second: Westbrook, Wade, Griffin, Anthony, Lopez
Third: Parker, Bryant, Ibaka, West, Duncan

Double counting PER:
First: Paul, Harden, LeBron, Durant, Lopez
Second: Westbrook, Wade, Griffin, Anthony, Duncan
Third: Parker, Bryant, Ibaka, West, Marc Gasol

Double counting WS/48:
First: Paul, Harden, LeBron, Durant, Marc Gasol
Second: Westbrook, Parker, Griffin, Anthony, Lopez
Third: Wade, Stephen Curry, Ibaka, Splitter, Chandler

lol fag.

[benji]

lol at u thinkin i did anything more than copy-paste the table from b-r then sort it in excel. took less time than u last in bed.

[Mandich]

took less time than u last in bed.

[Leftos]

benji, again, don't get why cube-rooting a multiplied value somehow makes things equally weighted. If the value before cube-rooting is the same, no matter what multiplicants you had, it's still going to be the same after you cube-root it. Unless you're doing something else other than cube-rooting the result of the multiplication that I didn't get from your message, it seems redundant.

All I'm saying is cube-root(x) == cube-root(y) if x == y. What am I missing? Care to elaborate on which step you apply the cube-root?

As for Game Score, it's just a simple metric that judges a player's performance in a game without per 36 normalization. So you take into account absolute production. Scaling over a season only means taking an average of what the Game Score for that player for all games he participated in that season. You can still easily turn it into Per 36 (and I've added such a metric to NST by default), but I'm not sure whether All-NBA Teams are supposed to be based on an absolute performance basis or take efficiency into account as well. In simple words, is a player that scores 30 points in 40 minutes per game worth more or less than a player who scores 20 points in 20 minutes? The general case says less, but are MPG taken into account when judging and creating the All-NBA teams?

[Patr1ck]

lol at u thinkin I did anything more than copy-paste the table from b-r then sort it in excel. took less time than u last in bed.

Wouldn't that be a low usage rate and a high efficiency?

[shadowgrin]

Gonna try to decipher benji's gibberish because it's a fun exercise in futility...

don't get why cube-rooting a multiplied value somehow makes things equally weighted.

Since he used three variables (PER, WS, WS/48) it made sense to get the 'root' of all three variables put together using a 'three' root.

If the value before cube-rooting is the same, no matter what multiplicants you had, it's still going to be the same after you cube-root it.
All I'm saying is cube-root(x) == cube-root(y) if x == y.

lol of course except the dope didn't use variables that are equal. PER != WS != WS/48

(abc)^1/3 = x
(a^1/3)(b^1/3)(c^1/3) = y
x = y