RealGM Top 100 List #2

Moderators: Doctor MJ, trex_8063, penbeast0, PaulieWal, Clyde Frazier

kayess
Sixth Man
Posts: 1,807
And1: 1,000
Joined: Sep 29, 2013

Re: RealGM NBA Top 100 List -- #2 

Post#241 » by kayess » Wed Jul 2, 2014 12:41 pm

penbeast0 wrote:
kayess wrote:I'm voting for Jabbar. Reasons were in thread 1, and despite leaning very close to Russell at one point, TrueLAFan's post sealed it for me. Will expound later as I want to make some comments about a couple of lines of thinking presented here.


Please expound as this is not a post sufficient to qualify your vote.


Vote: Kareem Abdul-Jabbar

I think his immense impact, year-on-year, over 20 years, trumps Russell's (for the sake of argument) superior impact, year-on-year, over 13 years; or yes, Kareem peaked high/close enough to Russell that his longevity gives him the edge.

He's already the best player in the league upon his arrival. One of the best single season turnarounds in NBA history, and the addition of Oscar takes them to the next level, winning a ring. Lose out to some great teams in the process, but again, it's about level of play, and that can't be questioned.

When he gets traded to the Lakers, it's like his Cavs LeBron days - just how far can one guy take a team of subpar rotation players? Far, if you're an all-history player, but you'll of course still lose to superior teams (especially if said superior team is being led by a player with an all-history peak).

Through this he maintains tremendous impact. He gets screwed out of a Finals MVP in 1980 (I don't blame the voters too much, it's hard not to get swayed by what is possibly the single greatest Finals game performance of all time), but was clearly the best player on that title team. TrueLAFan contextualizes Kareem's value in his later years better than I ever could, with the comparison to Tim Duncan, and his numbers with/without Magic all point to him still being a great, though obviously not as great, player in his decline. And everything else we know about him - his skillset, his fitness, his intelligence, etc., make this result unsurprising. When the Lakers needed a bucket in the half-court, they could always go to Kareem.

drza made some EXCELLENT points (that I feel will all eventually make the foundation for his impending epic KG post), but I have a few questions:

On historical trends pointing to the best offenses/defenses always being led by a perimeter player/big man, with certain schemes: To me, this doesn't necessarily scream "therefore there must be a ceiling for offenses led by big men" (the converse, "there must be a ceiling for defenses led by perimeter players", seems much easier to qualify, btw, but that's not what's being discussed here), but begs further investigation into how much credit the scheme/fit of the teammates add to a player's perceived impact. It's getting really late, so I'm not quite sure if this fit/portability has been accounted by ElGee's expected championship odds

Consider this:

Player A is bringing a +7 impact to a team that fits him moderately; in other situations with worse fit, he's a +7
Player B is bringing a +10 impact to a team that fits him perfectly; in other situations with worse fit, he's a +6

Given all that has been said about RAPM and all (I think it was Chicago76 saying that if it's +3 vs. +1, it quite clearly goes to the +3 player, but at higher levels even more rigorous analysis is needed), is it fair to give the fit/coaching/system a large portion of the credit, as well?

I hope I was able to articulate that with some clarity. I'm in this to learn anyways, so many great posts, please keep it coming!
lorak
Head Coach
Posts: 6,317
And1: 2,237
Joined: Nov 23, 2009

Re: RealGM NBA Top 100 List -- #2 

Post#242 » by lorak » Wed Jul 2, 2014 12:48 pm

ElGee wrote:realbig3's points have been making me think of something that I don't think I've seen brought up a lot.

Let's say you think Russell's peak defensive impact is worth 7 points. Even 8. And you look around the landscape today and don't see anybody over 4 or 5 on defense alone. (Personally, this is in line with my evaluations.) Consider that this is per game. And Russell's teams played a lot of possessions. Like, 20-25% more than we see today. Which means, on a per-possession basis, Russell would really be closer to a +6 today.


But only on defense. Overall his impact was lower, because he was negative offensive player - and it's consistent with Boston SRS, because even if we assume that rest of Celtics players were 0 (what is rather untrue, because it would means Hondo, S. Jones, Howell or even Cousy and Sharman are badly overrated), then Russell's impact wasn't bigger than around 6 SRS (maybe +8 in peak year), but in reality it was more like ~4-5 SRS. I wonder if anyone checked how Celtics were doing (SRS) in 1957 before and after Russell? BTW, it's really interesting that in 2nd and 3rd Russell's year Celtics were almost the same (still in terms of SRS) as in his rookie year, when he missed so much time because of Olympics.

My vote goes to KAJ, because I think his overall impact was bigger than Russell's. When we discussed Kareem's defense some time ago I discovered that he was better defensively than I was thinking. Not as good as really great defensive anchors, but good enough to make significant difference on that end of the floor and obviously he was more impactfull on D than Russell on offense (Bill was negative offensive player!), while their strengths (KAJ's offense, Russell's defense) are more or less on the same level if look only on impact relatively to era played. (however "on random team in NBA history" KAJ's offense would be more valuable than Russell's defense.)

drza wrote: That Kareem's impact was great...around a +7 SRS estimate...but that it was a step below the truly super elite of history.


Well, but drza, what was Russell's impact? Why you think it was bigger than 7? Because IMO data says Russell's impact on defense was at GOAT level (relatively to era played), but not overall impact, because Russell was negative offensive player, and thus his overall impact was lower than 7 SRS.

TrueLAfan wrote:I’m asking this not to be combative or rude, but as a genuine question—one that I’ve asked from time to time in player value discussions. If Oscar was as important and necessary as people say with the Bucks then … what about the rest? His first ten years in the league?

No matter how you slice it, Oscar's Cincinnati teams were .500 teams. (...) Where is the impact the Oscar is sometimes given credit for with the Bucks?


Where? Maybe here:

Royals with Oscar: 54.4 W/L%
Royals w/o Oscar: 22.2 W/L% (54 games missed)

Royals were crap organization, with crap roster, especially in era, when it wasn't possible to win without dominant big man and Royals didn't have such player (yes, I know who Lucas was or Embry or even Dierking...). It was bigger disaster than what KG had in Mnnesota or LeBron in Cleveland. Oscar leading them to 54 win% shows how GREAT he was, not how overrated he is.

I'm really surprised you used that kind of argument, when at the same time you are defending KAJ, who did not so well with Lakers before Magic arrived.
User avatar
eliasrapp98
Assistant Coach
Posts: 4,325
And1: 381
Joined: May 28, 2012
Location: Philly
       

Re: RealGM NBA Top 100 List -- #2 

Post#243 » by eliasrapp98 » Wed Jul 2, 2014 2:19 pm

My vote is for Kareem. Longevity, impact, titles, the skyhook. He dominated the NBA from black and white and NBA on NBC. He has everything an nba fan wants. If you want stats, he has it, if you want titles, he has it, if you want longevity, he has it, if you want MVPS/FMVP/etc, he has it.


Sent from my iPhone using Tapatalk
PG: Russell Westbrook, Reggie Jackson
SG: Andre Roberson, Anthony Morrow, Jeremy Lamb
SF: Kevin Durant, KJ McDaniels, Perry Jones
PF: Serge Ibaka, Nick Collison, Robert Covington
Cc: Al Jefferson, Steven Adams, Kendrick Perkins, Mitch McGary
Baller2014
Banned User
Posts: 2,049
And1: 519
Joined: May 22, 2014
Location: No further than the thickness of a shadow
     

Re: RealGM NBA Top 100 List -- #2 

Post#244 » by Baller2014 » Wed Jul 2, 2014 2:31 pm

This is such an incredibly close race, with a lot of good discussion. Sadly I'm unlikely to get a chance to post again on this thread before the time is up, so if anyone is curious I've added my vote count below (since others are doing it).

Kareem- 20 votes (Baller, Colts, RayBan, Imon, Greatness, Dr P, DannyN, Basketballefan, tlafan, trex, Ronny, MacGill, realbig3, Owly, Mutnt, Clyde, NArigo, Sacto, Kayess, Lorak)

Russell- 19 votes (Pen, fplii, Quinn, JBulls, HBkid, Dhodgkins, mmerl, batmana, Texas, Quo, Jaivl, Ardee, Dr MJ, D Nice, Drza, PCP, GC Pan, Rich316, Rico

Hakeem- 1 (90'salldecade- preferences Kareem though in his post)

I haven't included the following in my count, though they appear to express a preference, because the post was a little indecisive; UBfan (KAJ), Gregoire (Kareem or Wilt), Notnoob
colts18
Head Coach
Posts: 7,434
And1: 3,255
Joined: Jun 29, 2009

Re: RealGM NBA Top 100 List -- #2 

Post#245 » by colts18 » Wed Jul 2, 2014 2:51 pm

ElGee wrote:
This has come up before, but I think the 1960's Celtics won 11 titles because they were so much better than everyone else that their margin of error -- i.e. their effect of luck -- was great. The Celtics could withstand a 20% drop in play that coincided with a 20% increase from an opponent because they were already lapping them.

Now, how were they so dominant? Because Bill Russell was so dominant.

Obviously, people differ on how dominant. That's fair and perhaps the crux of the issue between GOAT and 3rd or 6th after a decade of +5 play vs. a decade of +8 play. But consider that the environment in the 9-team league was much more closely packed together than we see today. The standard deviations were smaller and the range among non-Celtics team much more compressed. Moving the needle 5 points was outstanding. West, Wilt, and yes Oscar were studs if they could take a -3 and make them a +3. (e.g. in 1963 West missed 25 games and LA was -2.1. With him they were 7.6 points better that year.)

From 1958-1969 Bill Russell missed 25 games where the opponent and Celtics were otherwise full strength. In those games, Boston was a -2.3 SRS team. In the other 943 games, they were +6.1. You don't need to believe that the sample perfectly captures Russell's value to see that there is some level of outlier-ness happening that gave Boston a freakish advantage over the entire league. Although you would have to think that that sample really undersold Boston's supporting cast to not think that Russell was catapulting a team to places no one else could take teams at the time.



Why did you not include 1957 in your assessment of Russell games missed? The Celtics had a better record without Russell in 1957 than with him. Do you think all of a sudden he turned from negative impact to super impact within 1 year? That is what the "impact" stats say. In 1991 Hakeem Olajuwon's team went 16-10 without him. Really good. If you went by your small sample size impact method you would say Hakeem wasn't much of an impact player. But if you went by 92-96 when his team went 7-27 without him you would say he is top 3 most impactful player. So which is it? Is 1991 a true indicator of Hakeem's impact or 92-96? Maybe the right answer is that you can't glean much from small sample sizes. I am surprised you are willing to use such a small sample size which comes out to 2 games missed per year. :lol:


Without Russell vs. overall Celtics:
26-26 W-L (.500)
0.70 SRS w/o vs. 5.38 SRS with (-4.68 SRS)

114.31 PPG vs. 110.71 PPG (+3.59)

112.77 PPG allowed vs. 104.77 PPG (-8.0)

Celtics played around ~120 possessions per game which means that Russell's impact was around +3.9 per 100. That's nothing special. Lots of players have surpassed that.

Russell was a defensive mastermind but those stats also say he was Kendrick Perkins like on offense.
ThaRegul8r
Head Coach
Posts: 6,448
And1: 3,037
Joined: Jan 12, 2006
   

Re: RealGM NBA Top 100 List -- #2 

Post#246 » by ThaRegul8r » Wed Jul 2, 2014 2:53 pm

lorak wrote:
TrueLAfan wrote:I’m asking this not to be combative or rude, but as a genuine question—one that I’ve asked from time to time in player value discussions. If Oscar was as important and necessary as people say with the Bucks then … what about the rest? His first ten years in the league?


[...]

I'm really surprised you used that kind of argument, when at the same time you are defending KAJ, who did not so well with Lakers before Magic arrived.


I actually found this odd myself, that a Kareem guy would do this, when considering one charge I've seen some people continually level against Kareem is "why didn't he elevate his teams from '74-75 to '78-79?" That's why I'm taking a closer look into it (missing the playoffs in '75 was his own fault for not being able to keep his temper under control—his hand injury was self-inflicted), but there is a double standard there.
I remember your posts from the RPOY project, you consistently brought it. Please continue to do so, sir. This board needs guys like you to counteract ... worthless posters


Retirement isn’t the end of the road, but just a turn in the road. – Unknown
colts18
Head Coach
Posts: 7,434
And1: 3,255
Joined: Jun 29, 2009

Re: RealGM NBA Top 100 List -- #2 

Post#247 » by colts18 » Wed Jul 2, 2014 3:06 pm

More notes on "impact" stats:

-In 2013, the Heat went 5-1 without LeBron
-Last season the Kings went 0-10 in the games that Demarcus Cousins missed

Were their impacts similar?
Notanoob
Analyst
Posts: 3,475
And1: 1,223
Joined: Jun 07, 2013

Re: RealGM NBA Top 100 List -- #2 

Post#248 » by Notanoob » Wed Jul 2, 2014 3:08 pm

I am leaning towards Kareem. I am still reading ThaRegul8r's topic on Kareem (it's quite long) and I have never been convinced that Russel's outsized defensive impact would translate to a similarly large impact in later years, especially if he was asked to do more offensively, if he even could. I don't want to cast my vote just yet, but it is quite likely that I'll vote for Kareem.
ElGee
Assistant Coach
Posts: 4,041
And1: 1,208
Joined: Mar 08, 2010
Contact:

Re: RealGM NBA Top 100 List -- #2 

Post#249 » by ElGee » Wed Jul 2, 2014 3:19 pm

Yes, I've looked at Russell's rookie year and posted in the Peaks Project all of the relevant WOWY information we have around his team:

Image

What you see is great impact for a rookie in 1957, especially one joining a team mid-season. It's not earth-shattering, but I think you clearly have a case of the 1957 Celtics being a better team than the year before (where they were already +1) and Russell not immediately having the impact he would have in ensuing, full seasons.

If you scan the rest of the decade, you'll see further corroborative evidence how just how valuable he was relative to other players -- Havlicek misses some games in 65 and 66 and Boston goes from a 57 to a 53 win team without him. Jones misses a bunch of games at the end of the decade, and while there's one blip in 1967 (also Russell's first year of coaching), from 66-69 the Celtics go from a 57-win team to 52-win team without Sam over the 4 years. The cumulative effect of Russell's missed games from 58-69 takes them from a 58-win team to a 34-win team.

And again, this is in a league without the 3-point shot, so the ability to create separation and impact the game is diminished. I don't think it's a coincidence that the top 5 SRS teams of the 2-point era came after expansion -- it was just harder to move the needle back then. You almost have to think of an 8 SRS in 1965 as a 13 SRS in the 90's.* And for me personally, that's a compelling argument to try and "normalize" someone's per-game impact in different eras.

*
Spoiler:
Standard deviation of teams in the 2-pt era was 3.9 SRS pts. In the 3-pt era 4.5 SRS pts. The difference was even greater pre-1967 expansion: 3.6 to 4.4, respectively.
Check out and discuss my book, now on Kindle! http://www.backpicks.com/thinking-basketball/
ElGee
Assistant Coach
Posts: 4,041
And1: 1,208
Joined: Mar 08, 2010
Contact:

Re: RealGM NBA Top 100 List -- #2 

Post#250 » by ElGee » Wed Jul 2, 2014 3:22 pm

Kareem's WOWY, for posterity:

Image
Check out and discuss my book, now on Kindle! http://www.backpicks.com/thinking-basketball/
penbeast0
Senior Mod - NBA Player Comparisons
Senior Mod - NBA Player Comparisons
Posts: 30,547
And1: 10,026
Joined: Aug 14, 2004
Location: South Florida
 

Re: RealGM NBA Top 100 List -- #2 

Post#251 » by penbeast0 » Wed Jul 2, 2014 3:30 pm

lorak wrote:
ElGee wrote:realbig3's points have been making me think of something that I don't think I've seen brought up a lot.

Let's say you think Russell's peak defensive impact is worth 7 points. Even 8. And you look around the landscape today and don't see anybody over 4 or 5 on defense alone. (Personally, this is in line with my evaluations.) Consider that this is per game. And Russell's teams played a lot of possessions. Like, 20-25% more than we see today. Which means, on a per-possession basis, Russell would really be closer to a +6 today.


But only on defense. Overall his impact was lower, because he was negative offensive player - and it's consistent with Boston SRS, because even if we assume that rest of Celtics players were 0 (what is rather untrue, because it would means Hondo, S. Jones, Howell or even Cousy and Sharman are badly overrated), then Russell's impact wasn't bigger than around 6 SRS (maybe +8 in peak year), but in reality it was more like ~4-5 SRS. I wonder if anyone checked how Celtics were doing (SRS) in 1957 before and after Russell? BTW, it's really interesting that in 2nd and 3rd Russell's year Celtics were almost the same (still in terms of SRS) as in his rookie year, when he missed so much time because of Olympics.

My vote goes to KAJ, because I think his overall impact was bigger than Russell's. When we discussed Kareem's defense some time ago I discovered that he was better defensively than I was thinking. Not as good as really great defensive anchors, but good enough to make significant difference on that end of the floor and obviously he was more impactfull on D than Russell on offense (Bill was negative offensive player!), while their strengths (KAJ's offense, Russell's defense) are more or less on the same level if look only on impact relatively to era played. (however "on random team in NBA history" KAJ's offense would be more valuable than Russell's defense.)

drza wrote: That Kareem's impact was great...around a +7 SRS estimate...but that it was a step below the truly super elite of history.


Well, but drza, what was Russell's impact? Why you think it was bigger than 7? Because IMO data says Russell's impact on defense was at GOAT level (relatively to era played), but not overall impact, because Russell was negative offensive player, and thus his overall impact was lower than 7 SRS.

TrueLAfan wrote:I’m asking this not to be combative or rude, but as a genuine question—one that I’ve asked from time to time in player value discussions. If Oscar was as important and necessary as people say with the Bucks then … what about the rest? His first ten years in the league?

No matter how you slice it, Oscar's Cincinnati teams were .500 teams. (...) Where is the impact the Oscar is sometimes given credit for with the Bucks?


Where? Maybe here:

Royals with Oscar: 54.4 W/L%
Royals w/o Oscar: 22.2 W/L% (54 games missed)

Royals were crap organization, with crap roster, especially in era, when it wasn't possible to win without dominant big man and Royals didn't have such player (yes, I know who Lucas was or Embry or even Dierking...). It was bigger disaster than what KG had in Mnnesota or LeBron in Cleveland. Oscar leading them to 54 win% shows how GREAT he was, not how overrated he is.

I'm really surprised you used that kind of argument, when at the same time you are defending KAJ, who did not so well with Lakers before Magic arrived.


Have you done this same analysis for Lucas who is a strong stat guy? I've always wondered how Cincinnati can have two players as outstanding as Oscar and Lucas and still be so suck. Either (a)Oscar's impact is not that great, (b)Lucas's impact is not that great, (c)the rest of the team is THAT far below average (this would include the coaching). I'd love to see it if you have it.
“Most people use statistics like a drunk man uses a lamppost; more for support than illumination,” Andrew Lang.
User avatar
Texas Chuck
Senior Mod - NBA TnT Forum
Senior Mod - NBA TnT Forum
Posts: 92,791
And1: 99,364
Joined: May 19, 2012
Location: Purgatory
   

Re: RealGM NBA Top 100 List -- #2 

Post#252 » by Texas Chuck » Wed Jul 2, 2014 3:38 pm

colts18 wrote:More notes on "impact" stats:

-In 2013, the Heat went 5-1 without LeBron
-Last season the Kings went 0-10 in the games that Demarcus Cousins missed

Were their impacts similar?


A championship-level team can often survive the loss of a key player for a game or two especially when you have a couple guys in Wade and Bosh who are more than capable of stepping up and providing some of what is lost from Lebron's absence. The Kings were a highly dysfunctional team with a losing culture missing their key guy.

One's importance to one's team does not linearly equal impact
ThunderBolt wrote:I’m going to let some of you in on a little secret I learned on realgm. If you don’t like a thread, not only do you not have to comment but you don’t even have to open it and read it. You’re welcome.
ardee
RealGM
Posts: 15,320
And1: 5,397
Joined: Nov 16, 2011

Re: RealGM NBA Top 100 List -- #2 

Post#253 » by ardee » Wed Jul 2, 2014 3:39 pm

ElGee wrote:Kareem's WOWY, for posterity:

Image


ElGee, you're the most informed stat guy here. So in your opinion, what is a big enough sample size to take a single season WOWY seriously?

I remember studying sample sizes in a stats class but I can't quite remember the formula to check if it's big enough haha.
ElGee
Assistant Coach
Posts: 4,041
And1: 1,208
Joined: Mar 08, 2010
Contact:

Re: RealGM NBA Top 100 List -- #2 

Post#254 » by ElGee » Wed Jul 2, 2014 3:45 pm

penbeast0 wrote:
lorak wrote:
ElGee wrote:realbig3's points have been making me think of something that I don't think I've seen brought up a lot.

Let's say you think Russell's peak defensive impact is worth 7 points. Even 8. And you look around the landscape today and don't see anybody over 4 or 5 on defense alone. (Personally, this is in line with my evaluations.) Consider that this is per game. And Russell's teams played a lot of possessions. Like, 20-25% more than we see today. Which means, on a per-possession basis, Russell would really be closer to a +6 today.


But only on defense. Overall his impact was lower, because he was negative offensive player - and it's consistent with Boston SRS, because even if we assume that rest of Celtics players were 0 (what is rather untrue, because it would means Hondo, S. Jones, Howell or even Cousy and Sharman are badly overrated), then Russell's impact wasn't bigger than around 6 SRS (maybe +8 in peak year), but in reality it was more like ~4-5 SRS. I wonder if anyone checked how Celtics were doing (SRS) in 1957 before and after Russell? BTW, it's really interesting that in 2nd and 3rd Russell's year Celtics were almost the same (still in terms of SRS) as in his rookie year, when he missed so much time because of Olympics.

My vote goes to KAJ, because I think his overall impact was bigger than Russell's. When we discussed Kareem's defense some time ago I discovered that he was better defensively than I was thinking. Not as good as really great defensive anchors, but good enough to make significant difference on that end of the floor and obviously he was more impactfull on D than Russell on offense (Bill was negative offensive player!), while their strengths (KAJ's offense, Russell's defense) are more or less on the same level if look only on impact relatively to era played. (however "on random team in NBA history" KAJ's offense would be more valuable than Russell's defense.)

drza wrote: That Kareem's impact was great...around a +7 SRS estimate...but that it was a step below the truly super elite of history.


Well, but drza, what was Russell's impact? Why you think it was bigger than 7? Because IMO data says Russell's impact on defense was at GOAT level (relatively to era played), but not overall impact, because Russell was negative offensive player, and thus his overall impact was lower than 7 SRS.

TrueLAfan wrote:I’m asking this not to be combative or rude, but as a genuine question—one that I’ve asked from time to time in player value discussions. If Oscar was as important and necessary as people say with the Bucks then … what about the rest? His first ten years in the league?

No matter how you slice it, Oscar's Cincinnati teams were .500 teams. (...) Where is the impact the Oscar is sometimes given credit for with the Bucks?


Where? Maybe here:

Royals with Oscar: 54.4 W/L%
Royals w/o Oscar: 22.2 W/L% (54 games missed)

Royals were crap organization, with crap roster, especially in era, when it wasn't possible to win without dominant big man and Royals didn't have such player (yes, I know who Lucas was or Embry or even Dierking...). It was bigger disaster than what KG had in Mnnesota or LeBron in Cleveland. Oscar leading them to 54 win% shows how GREAT he was, not how overrated he is.

I'm really surprised you used that kind of argument, when at the same time you are defending KAJ, who did not so well with Lakers before Magic arrived.


Have you done this same analysis for Lucas who is a strong stat guy? I've always wondered how Cincinnati can have two players as outstanding as Oscar and Lucas and still be so suck. Either (a)Oscar's impact is not that great, (b)Lucas's impact is not that great, (c)the rest of the team is THAT far below average (this would include the coaching). I'd love to see it if you have it.


In 65 Lucas missed 14 games controlling for Oscar.

65 Cin WITH Lucas: +2.2 SRS
65 Cin W/OUT Lucas: +4.1 SRS

In 70 Lucas missed 17 games controlling for Nate Thurmond.

70 SFW WITH Lucas: -3.0 SRS
70 SFW w/OUT Lucas: -0.3 SRS

I never did 1969, a year in which he missed 8 games...
Check out and discuss my book, now on Kindle! http://www.backpicks.com/thinking-basketball/
penbeast0
Senior Mod - NBA Player Comparisons
Senior Mod - NBA Player Comparisons
Posts: 30,547
And1: 10,026
Joined: Aug 14, 2004
Location: South Florida
 

Re: RealGM NBA Top 100 List -- #2 

Post#255 » by penbeast0 » Wed Jul 2, 2014 3:47 pm

Doesn't look good for Jerry Lucas
“Most people use statistics like a drunk man uses a lamppost; more for support than illumination,” Andrew Lang.
ThaRegul8r
Head Coach
Posts: 6,448
And1: 3,037
Joined: Jan 12, 2006
   

Re: RealGM NBA Top 100 List -- #2 

Post#256 » by ThaRegul8r » Wed Jul 2, 2014 3:47 pm

ElGee wrote:Yes, I've looked at Russell's rookie year and posted in the Peaks Project all of the relevant WOWY information we have around his team:

Image

What you see is great impact for a rookie in 1957, especially one joining a team mid-season. It's not earth-shattering, but I think you clearly have a case of the 1957 Celtics being a better team than the year before (where they were already +1) and Russell not immediately having the impact he would have in ensuing, full seasons.

If you scan the rest of the decade, you'll see further corroborative evidence how just how valuable he was relative to other players -- Havlicek misses some games in 65 and 66 and Boston goes from a 57 to a 53 win team without him. Jones misses a bunch of games at the end of the decade, and while there's one blip in 1967 (also Russell's first year of coaching), from 66-69 the Celtics go from a 57-win team to 52-win team without Sam over the 4 years. The cumulative effect of Russell's missed games from 58-69 takes them from a 58-win team to a 34-win team.

And again, this is in a league without the 3-point shot, so the ability to create separation and impact the game is diminished. I don't think it's a coincidence that the top 5 SRS teams of the 2-point era came after expansion -- it was just harder to move the needle back then. You almost have to think of an 8 SRS in 1965 as a 13 SRS in the 90's.* And for me personally, that's a compelling argument to try and "normalize" someone's per-game impact in different eras.

*
Spoiler:
Standard deviation of teams in the 2-pt era was 3.9 SRS pts. In the 3-pt era 4.5 SRS pts. The difference was even greater pre-1967 expansion: 3.6 to 4.4, respectively. If we removed the Celtics, the


The spoiler is incomplete. It cuts off at: "If we removed the Celtics, the—"
I remember your posts from the RPOY project, you consistently brought it. Please continue to do so, sir. This board needs guys like you to counteract ... worthless posters


Retirement isn’t the end of the road, but just a turn in the road. – Unknown
ardee
RealGM
Posts: 15,320
And1: 5,397
Joined: Nov 16, 2011

Re: RealGM NBA Top 100 List -- #2 

Post#257 » by ardee » Wed Jul 2, 2014 3:50 pm

penbeast0 wrote:Doesn't look good for Jerry Lucas


Yeah it makes sense... I've read that Lucas was the most stat-obsessed guy in the league back then. Makes sense, considering he's the only player ever minus Wilt to average a 20/20 and he's not a third of the natural talent Wilt was.
ElGee
Assistant Coach
Posts: 4,041
And1: 1,208
Joined: Mar 08, 2010
Contact:

Re: RealGM NBA Top 100 List -- #2 

Post#258 » by ElGee » Wed Jul 2, 2014 4:05 pm

ardee wrote:
ElGee wrote:Kareem's WOWY, for posterity:

Image


ElGee, you're the most informed stat guy here. So in your opinion, what is a big enough sample size to take a single season WOWY seriously?

I remember studying sample sizes in a stats class but I can't quite remember the formula to check if it's big enough haha.


I actually analyzed the variance in WOWY a few years ago and wrote about it. In general, a sufficient sample size comes from a sufficient reduction in variance. That's the relationship. I use the following guidelines when scanning WOWY:

< 7 or 8 games -- could almost be pure noise
8-12 games -- noisy but some probability of accuracy
12-20 games -- nice size, but still subject to some noise
> 20 games -- here is where we really see stability.

I'll caution that > 20 game samples are not 100% reflective, or even 99.9%, but that's when you enter into that 90-95% zone that scientists love to arbitrarily call "significant."

So in general, I'm not paying much attention to a single 5-game sample, I'm gleaning at about 8-20 games, and I'm very interested at over 20 games.

As a PS, you have to judge how many moving parts are involved when you increase the sample across 2 seasons (or 3 as I have done in the past) or involve trades. With Russell's sample, if it were 1 year, 26 games, I'd find that compelling evidence of his "real" value on that team. Across 11 years, I think it's a large enough sample to tell us something ("Hey, what REALLY happened when Russ didn't play over the years?") but spread out enough that I would not look at it as reflective of his actual impact in Boston over the years. It could be perfectly accurate, but we can't be confident enough from that kind of sample.
Check out and discuss my book, now on Kindle! http://www.backpicks.com/thinking-basketball/
Dr Spaceman
General Manager
Posts: 8,575
And1: 11,211
Joined: Jan 16, 2013
   

Re: RealGM NBA Top 100 List -- #2 

Post#259 » by Dr Spaceman » Wed Jul 2, 2014 4:12 pm

First I want to apologize, I haven't participated as much as I would've liked so far. But discussions have been fantastic, and I've been having a great time reading them.

Anyway, as with #1, my vote here is: Bill Russell

Essentially, any player in a debate with Kareem is going to come down to who's career is more "valuable", as his longevity is enough to give him a run at GOAT. Like has been pointed out here, Kareem gives you 5 extra years of very good play in addition to all of his years as an elite player. I take longevity very seriously, as will become evident as the project pushes forward. So it is no easy task for me to place anyone over Kareem (at one point before the project I very seriously thought about going with him as GOAT). The trinity of MJ, Russell, and KAJ is so close to me in part because they all bring vastly different skills to the table, and impact the game in contrasting ways.

With all that said, I value Russell's defense more than anything any other player has brought to the table. That alone is not enough to put him over Kareem and his two-way difference making, but here's the one thing that will always differentiate Russell from the rest to me: His prime lasted basically 13 years,longer and more sustained than really any other player. He adapted to a volatile and expanding game, and remained the best pretty much throughout. And I'm of the opinion that if player is an absurd outlier in one area (as Bill's defense certainly is) then it outweighs other players' lesser but more varied contributions. In that sense, Russell is arguably the most valuable player in NBA history, as his defense is so far above everyone else's.

Era based arguments sort of fall apart to me when I consider that throughout much of his career, it was the NBA trying to adapt to him, rather than the other way around. We're talking about the smartest player to ever play, who was so far ahead of everyone else that he was able to dominate in a way we've never seen before or since. I don't think transporting him forward in time changes that.
“I’m not the fastest guy on the court, but I can dictate when the race begins.”
TrueLAfan
Senior Mod - Clippers
Senior Mod - Clippers
Posts: 8,267
And1: 1,795
Joined: Apr 11, 2001

Re: RealGM NBA Top 100 List -- #2 

Post#260 » by TrueLAfan » Wed Jul 2, 2014 4:14 pm

Lucas was really weird. Eidetic. Off personality. Said he “never really wanted to be a pro basketball player.” He had bad knees and played huge minutes. But Lucas has the Oscar thing going for him (irony alert!). When he went to a great team, he elevated them and put them over the top. When he went to the Knicks, he was a Godsend … Reed had been hurt, and it was Willis Reed for Christ’s sake. Lucas is a fundamentally different player than Reed, and it took the Knicks a while to figure it out—they were 6-9 to start, the season, closed with 42-25 and made it to the finals. They won the title the next year with Lucas playing a smaller but still (very) important part. You don’t hear (nearly) as much about Lucas as you do Oscar in terms of his role on a championship team … but Lucas was very important in much the same way in a contemporary context.

Not 100% sold on the On/Off numbers. This is a subjective opinion. Lucas, despite his big numbers, was definitely Robin. Oscar usage rate must have been astronomical—22 shot attempts, 10 apg, and going to line 11 times a game? Jesus. It’s harder to replace that when it goes down, IMO—especially in the short term . The Royals were a bad organization, but they generally had well crafted teams. Compared to the Lakers, they maintained a balance of offense and defense, big and small. You got perimeter shooting (Twyman, Smith, Van Arsdale), good D and adequate scoring from the perimeter and one half of the PF/C position (Embry, Boozer, Hawkins, Hairston, Love). I disagree that that the Royals frontline was an area of weakness. It seems like adding an inside/outside combo like Oscar and Lucas to that would help out more than it did. But the bottom line is results, and those teams didn’t produce despite having (IMO) some balance and quality. The mid-70s Lakers, for instance, had no frontcourt players other than Kareem, and marginal perimeter D. If the three had been around, they might have been able to make a run like the D12 Magic. Without it, they were a 45-50 win team. But they were a team with flaws and gaps. I don’t see that as much with the Royals. Never have. But I can how losing such a massively ball-dominant player as Oscar, a guy that did so much in so many different areas, must have been devastating.
Image

Return to Player Comparisons