RealGM Top 100 List 2017 -- #13

[ceiling raiser]

Here's 01-06, I think it's all I have:

http://www8.zippyshare.com/v/qg0gCAUK/file.html

Thanks!

No problem, I think this is the raw data in case you need it:

http://www111.zippyshare.com/v/0z50Kb58/file.html

[eminence]

See fpliii has already got it covered, but I think I had a link for 07-08 that still worked from basketballvalue

http://www.basketballvalue.com/publicdata/matchups20072008reg20081211.zip (should download the zip file)

[penbeast0]

Some Regression Results...

Any explanation for Marc Gasol being third all time over the likes of Jordan, Garnett, etc.? I'd have been less surprised at Pau . . .

[Outside]

West over Oscar - Oscar's perfectionist style may have been grating on chemistry

Do explain. One thing I gained from Jerry West recent interviews is that he was continuously depressed and anxious during his playing days. He shut off the world around him and lived through basketball. I wonder whether any of this filtered into how he interacted with his teammates? We don't hear much drama between West and Baylor tho.

I was also going to highlight this part of Dr Positivity's post because I've heard this type of thing repeatedly from players of his era in books and articles. My impression is that Oscar was not only a perfectionist but demanded that you do things precisely the way he wanted them done, which can lead to less than ideal chemistry, especially when Big O bites your head off when you don't do what he wants. There is often more than one good option for a player to take in a given situation, such as pick and pop vs pick and roll, but even if Oscar was right in his criticism, which he probably was most of the time, it's a grating way to play a team game that would have some teammates thinking more about what does Oscar want me to do here instead of just playing.

West's depression and anxiety came from a much different place. By all accounts that I've seen or read, he was a great teammate, well respected by teammates and opponents alike. But he was driven by winning, or, perhaps more accurately, haunted by losing. So many heartbreaking losses in the playoffs were a terrible burden that ate at him. Even as GM for the Lakers during the Shaq-Kobe three-peat, he couldn't bring himself to watch the games live in the arena. But as a player, none of that impacted his performance or caused him to lash out at teammates during those playoff battles. He was a rock they could depend on during those times. It wasn't until after the games, after the playoffs, that he went to that dark place.

Magic did much the same after losing to the Celtics in 1984, but he got the redemption of beating the Celtics in 85 and 87. West did eventually get a title, but never over the Celtics, and Magic might've ended up the same if he'd been 1-8 in finals and never beaten the Celtics instead of being 5-4 in the finals and 2-1 over Boston.

To me, West was all about winning, while Oscar was more about playing the right way, which had the side effect of sometimes getting in the way of building the team chemistry necessary for winning.

If I get the time, I'll try to find quotes from players to back up that take.

[scrabbarista]

13. Karl Malone

14. Julius Erving

I. Karl Malone Bob Pettit are tied among remaining players in my MVP voting metric. Julius Erving is next after them.

II. Malone is second among remaining players in my "Honors" metric, after Jerry West.

III. Malone is the only player, along with Kareem, to have over 60,000 career regular season points, rebounds, assists, steals, and blocks. Kevin Garnett, in fourth all-time, has 50,074. This gigantic number for Malone is a major part of his case for being so high despite his relatively "lackluster" playoff resume.

IV. Speaking of the playoffs, Malone is 9th all-time in playoff points, rebounds, assists, blocks, and steals. He is above all remaining players, as well as Bill Russell, Larry Bird(!), Hakeem Olajuwon, and Kevin Garnett.

V. Significant parts of Erving's case come from his time in the ABA, which I penalize pretty heavily (I think I have the penalty at minus 30% for most accomplishments). Even with that penalty, his numbers are impressive, as is the fact that he was the best player on two championship teams while in his prime (also penalized at 30%). The only players left who have that on their resumes are Mikan, Isiah Thomas, and Dave Cowens, none of whom I would personally consider this high, meaning he is unique among players in consideration.

VI. Even with a 30% penalty for his ABA numbers, Erving is 12th all-time in postseason points, rebounds, blocks, steals, and assists. He finishes above Hakeem Olajuwon, and with over 38% more than Kevin Garnett (6800 to 4900 - that's with a 30% penalty for ABA totals).

[Pablo Novi]

As for all these players being from the last 5 years...

From the top 12...

2 are from the 60s (Russell and Wilt)
1 is from the 70s and 80s (Kareem)
2 are from the 80s (Magic and Bird)
2 are from the 80s and 90s (Jordan and Hakeem)
1 is from the 90s and 00s (Shaq)
3 are from the 00s (Duncan, Kobe, and Garnett)
1 is from the 00s and 10s (LeBron)

That's a pretty even distribution, and the discussion is between a 50s guy (Mikan), a 00s guy (Dirk), two 70s and 80s guys (Moses and Dr. J), two 60s guys (Oscar and West), and two 90s guys (Karl and D-Rob). Maybe Barkley as well, who would also fit into that 80s/90s era.

Not to mention, basketball just gets better over the years. Makes sense that the later generations have more all timers. There's just more talent to go around.

But again, that's not even what we're seeing here. It's a really even distribution. But yeah, guys like Wade and CP3 and Nash SHOULD probably be part of the conversation soon, along with guys like Stockton, Ewing, and Pippen. I think it's too early to discuss KD or Curry in the top 20-30 though.

I call what you refer to here as "Equality of 'Decades'"; and I love basically everything you say here.

For me, there's three "Equalities" that are important to me: "Equality of All-Time Greats In Both Leagues in Dual-League years" (NBL vs BAA; ABA vs NBA (after the first 2-3 years); "Equality of Positions" (in each descending set of 5 GOAT spots, I have one player from each position; though Centers pretty much lead each set of five); and "Equality of 'Decades'", vis-à-vis the All-Time Greats, is a huge thing.

So I find it very pleasing that we've COLLECTIVELY reflected that. At the same time, for me, the DEPTH, over-all quality, of the level of play keeps rising; so I've as I go lower into my GOAT list, the more recent decades start gaining an "advantage" number-of-players-wise.

My vote:
1st: Dr J.
2nd: the Big "O"
After that, Karl Malone
I have all three of these guys, in this order, in my GOAT #s 6-10, as Dr J (GOAT #7), "O" (GOAT #9), Karl (GOAT #10).
(I have Wilt as my GOAT #6 and Kobe as my GOAT #8.)

My GOAT #s 11-15 are: Shaq, Jerry West, Bird, Cousy, and Pettit.

I do have a "technical" question about this voting process. Last vote I felt strongly enough that Karl Malone deserved to be taken before KG, that I switched from voting Dr J first and "O" second; to voting Karl first and Dr J second.

So, I guess I'm asking for "guidance" on this type of "agenda-driven" voting.

Meanwhile, the most-basic criteria I use for judging for my GOAT rankings is each's number of ALL-League (ALL-NBA, ALL-ABA, ALL-NBL) 1st-Team, then 2nd-Team, then 3rd-Team selections - for me, nothing trumps the considered COLLECTIVE opinion of the very people whose job it is to report on our sport as it unfolds.

In my next post I'll include a copy of the results of "Points" system I developed to try to accurately reflect that ALL-League selection process:

[penbeast0]

...
I do have a "technical" question about this voting process. Last vote I felt strongly enough that Karl Malone deserved to be taken before KG, that I switched from voting Dr J first and "O" second; to voting Karl first and Dr J second.

So, I guess I'm asking for "guidance" on this type of "agenda-driven" voting....

Not only is it against the spirit of the project but note that if you vote Erving first and Karl Malone second, if Erving doesn't win your vote STILL counts for Karl Malone . . . unless it comes down to Erving v. Malone in which case the guy you didn't feel deserved it would win and that would suck. That's why we have the two part voting system.

[Blackmill]

Some Regression Results...

Any explanation for Marc Gasol being third all time over the likes of Jordan, Garnett, etc.? I'd have been less surprised at Pau . . .

For starters, players like Jordan and Garnett have several years in the regression when they were well past their prime. That probably reduces their fitted values.

It's generally hard to discern what a regression is "thinking". Sometimes you can find clear reasons. For instance, including the 90's in any regression to ascertain Duncan's impact will produce a diminished value, since Robinson was so good for so many years, but wasn't the same player after '98. Otherwise, when you see a player who has an unexpectedly high value, it's because the player's teammates have unexpectedly low values. Why the teammates have such low values can sometimes be explained by the years included in the sample but it's not necessarily obvious. This is convoluted by how regressions over multiple years can yield very different results than from averaging the regression results of each individual year.

Observe Marc Gasol's fitted values over his career:

09: -0.540940581623
10: 2.23989228559
11: 3.09267210586
12: 1.30040736305
13: 2.57433348528
14: 1.17990868241
15: 0.708887324446
16: 0.18248724379
17: 1.21970340606

Compare his single year results to the 09-17 regression:

09-17: 4.68932855202

You can see the full regression produces a significantly higher value than any single year. Why? I don't know. I plan to experiment with more powerful models than standard regression. I'll also see about implementing a model that produces lower values for players it's uncertain of. I won't dwell on Marc's rating too much unless he shows up just as strong in future iterations.

[BasketballFan7]

1. Julius Erving
2. Dirk Nowitzki

My argument for Erving is below.

I'll add a bit on Erving vs. West because West is gaining momentum. I don't put a ton of emphasis on longevity but it is pertinent here.
1. Neither West nor Erving's totals are inflated by low-impact, nigh irrelevant seasons/games/minutes. Neither player took time to adjust to the NBA. Nor did they linger too long past their primes.

2. Playoff availability is a big deal.

Games Played
West: 1,085
Erving: 1,432
Difference: 347

Minutes Played:
West: 42,892
Erving: 52,759
Difference: 9,867

Massive differences here that amount to the equivalent of ~3 seasons of play at the minimum. Very possibly 4+ seasons when health is taken into context.

Playoff Availability:

I'm not sure to what extent each individual poster values playoff availability. To me, this is a very big deal. I don't have a chance of winning a championship, ie accomplishing my goal, if my star is out for the playoffs.

- West missed the 1971 playoffs due to a knee injury
- West missed the 1974 playoffs due to a groin injury

Erving and his ABA peak:

I'll post here what I reiterate below in one of my quotes. 1976 Erving dominated two future NBA teams during the 1976 ABA playoffs, including Bobby Jones, David Thompson, and Dan Issel led Nuggets that had the NBA's best defense in 1977 (by a decent margin) and the NBA's second best expected W-L. Erving averaged ~38 PPG, 6 APG, 14 RPG against Denver. The other team that Erving played in the 1976 playoffs was the Spurs, who also transitioned successfully to the NBA (44-38 record in 1976).

In 1977, Erving's statistics fell off due to poor roster construction. He nevertheless increased his volume in the playoffs. His season culminated in his virtuoso performance in the finals against the Bill Walton-led Trailblazers, where Erving averaged 30 - 7 - 5 on 21.5 field goal attempts per game, including 40-8-6 in the game 6 that they lost by 2 and were eliminated.

My argument is not that Erving's peak is based wholly on these playoff series, although I do value playoff performance. These series more-so serve to add validity to his 1976 peak.

My post + previous argument

Vote: Julius Erving
Alternate: Kobe Bryant

These guys score closely in career value to me. Erving comes out a bit ahead across the board. He had a higher peak and by my criteria superior longevity. Kobe played more minutes but a substantial chunk of those minutes came post-2012 and pre-2000, and those seasons don't add a ton of value for me (2013 would have added significant value, enough to push him over Erving, had he not been injured late in the season, causing him to miss the playoffs). So, for me, longevity is a push at best and more likely going to Erving. Both players showed the ability to play with other star talent (that said, roster construction as a whole was far superior for Kobe) and both players tend to be overrated defensively.

Erving tends to be underrated altogether IMO. His reputation is such that one may come to believe that he couldn't handle the ball, shoot, or really do anything outside of be athletic and finish at the rim. This likely has to do with his PPG dropping upon entering the NBA. To me, that's on roster construction. You aren't going to convince me that 77-79 Erving was somehow significantly inferior to the 1980 and 1981 MVP winning version.

Lastly, I prefer Erving's less abrasive personality. I have Magic and Bird/Dirk after this pair.

I just deleted my response on accident

To be brief-

Erving's 1977 playoff run adds validity to his performance. He dominated the Nuggets and the Spurs, two teams that would transition to the NBA the following season with significant roster carryover and achieve 50 and 44 wins, respectively. In particular, the Nuggets team he faced in the finals was excellent by both ABA and NBA standards. The following year the 1977 Nuggets had the best defense in the NBA, as well as the second best expected win-loss. I certainly believe Erving's 1976 to be an upper-echalon quality season.

His NBA production was stymied by poor roster construction. This affected his averages. During the 1977 playoffs he raised his game once again, both overall and in particular in the finals, where he averaged 30-7-5 against Portland after only putting up 21.6 PPG in the regular season. Portland was obviously an excellent team.

I don't have much time to go more in depth, particularly after already having to re-write this

Skill-set wise... I don't put much emphasis into that here. To be clear, Kobe had marvelous versatility. IMO this is valuable because it allowed him to maintain production against damn-near any defense that opposed him. But Erving was so good at what he did do that the inferior versatility doesn't bother me. I find his versatility to be underrated as is. Era-relative, which is all I care for (I have Russell at number one), his handle didn't impede him. You don't do what he did as a slasher without a handle. And he wasn't a non-shooter or non-passer, at least not to the extent where it hindered him.

Kobe's skill-set distinguishes him against players who I feel could be limited against `playoff defenses. For instance, I have Kobe over Malone and Bird (although Bird's playoff drop-off obviously wasn't likely due to versatility issues). I don't feel that Erving had an issue in this regard.

And a related argument:

But Dr J was elite ballhandler for his position. West wasn't. You don't understand, even with if MJ was a better ballhandler (he was elite so there is nothing strange with that), he handled the ball in illegal way. He couldn't palm the ball in 1970s. Even rookie MJ played in an era when refs started allowing more flashy dribble moves. It would be easier for Julius Erving to handle the ball today. Now everyone carries the ball. Without carrying rules it's so easy that even bigs try to be flashy ballhandlers.

Last point, bolded part is just not true. Julius played in the same league with MJ and he was productive even in his last season. People overrate handling abilities. Elgin Baylor didn't use many dribble moves and he could beat any defender off the dribble. All this behind the back and between the legs stuff isn't really important for basketball player. If you think it is, Dr J was quite flashy ballhandler for his era. Sometimes he was allowed to do more with the ball than the rest because people loved him. He could do all important things to beat defender off the dribble and he could beat double teams with his dribbling. Players now are more comfortable with his dribbling because they basically can't do any illegal dribbles with the way refs call the game. They can also travel in almost any possesion.

If you wish, I can break down Dr J handling ability with video evidence.

PockyCandy's video:

Watch on YouTube

Most of the first 2/3 of the video is near the rim action. However, there are a few instances where he exhibits his ball handling by going between the legs or using a spin move.

The last three minutes show jump shots, passes, and blocks. The game commentator on at least two occasions references Erving's shooting ability.

ABA footage is very rare, so this video's content was drawn from a restricted number of games. Despite that, it demonstrates that Erving was more diverse than what common perception leads one to believe.

[Winsome Gerbil]

13) Mailman
14) Oscar

And actually, I am not sure I can add much that either I haven't said, or that a fellow poster didn't say in the postmortem of the #12 thread:

Karl Malone

His averages for 16 years between 1986-2003

22/10/2
28/12/2
29/11/3
31/11/3
29/12/3
28/11/3
27/11/4
25/12/4
27/11/4
26/10/4
27/10/4
27/10/5
24/9/4
26/10/4
23/8/5
22/9/4
21/8/5


In the 11 straight bolded seasons from 1989-1999 he was all nba 1st team.

If you want to talk about player durability look no further than Karl Malone.

In 17 out of his 19 seasons he played 80 or more games
For comparison Kevin Garnett has only played 80 games 8 out of the 19 years of his career
Not only that but Malone accomplished this whiles playing 2,624 more playoff minutes than Garnett in his career

He's the oldest player in Nba history to win Mvp at age 35.


His longevity in the playoffs is just as impressive

22/8/1
20/10/1
30/12/2
31/16/1
25/10/2
30/13/3
29/11/3
24/10/2
27/12/3
30/13/4
27/10/4
26/11/3
26/10/3
22/11/5
27/9/3
28/9/3
20/8/5
20/7/4

That's an 18 year span in the post season

So if you guys are criticizing Magic and Bird for their durability and longevity, that same focus should be propelling Malone up in these rankings.


Also for you guys who put great value in the "player efficency stat" or "PER"rating, Malone's consistency in that statistic speaks for itself.

For 13 seasons from 1988-2001 Karl Malone was top 5 in "PER" in a league that included extremely efficient players like ; David Robinson, Shaq, Barkley, Ewing, Hakeem and Jordan all in their prime

1988-1989 - #5 In PER (24.4)
1989-1990- #2 In PER (27.2)
1990-1991- #5 In Per (24.8)
1991-1992- #3 in PER (25.4)
1992-1993- #3 in PER (26.2)
1993-1994- #5 in PER (22.9)
1994-1995- #5 In Per (25.1)
1995-1996- #4 in PER (26.0)
1996-1997- #1 in Per (28.9)
1997-1998- #2 in Per (27.9)
1998-1999- #2 in Per (25.6)
1999-2000- #2 in PER (27.1)
2000-2001- #4 In Per (24.7)

That shows ridiculous efficiency over such a long period of time. That very few can match

Also for you guys who seem to value win shares as a legitimate stat.. Malone was consistently elite in that as well


1988-1989 - #5 in Win Shares (15.2)
1989-1990- #4 in Win Shares (15.9)
1990-1991- #3 in Win Shares (15.5)
1991-1992- #2 in Win Shares (15.1)
1992-1993- #3 in Win Shares (15.4)
1993-1994- #4 In Win Shares (13.4)
1994-1995- #4 in Win Shares (13.8)
1995-1996- #3 in Win Shares (15.1)
1996-1997- #2 in Win Shares (16.7)
1997-1998- #1 in Win Shares (16.4)
1998-1999- #1 in Win Shares (9.6)
1999-2000- #2 in Win Shares (15.3)
2000-2001- #5 in Win Shares (13.1)

That again is 13 straight years of being top 5 in the league in Win Shares.

In Offensive win Shares he's 6th all time trailing only Jordan, Oscar, Kareem, Wilt and Stockton.

In Defensive win shares he's a 6th all time (Ahead of Garnett) only trailing Russell, Duncan,Kareem , Hakeem and Wilt

For his career he only trails Wilt Chamberlin and Kareem in total win shares, he's 3rd all time.

So for a 13 year span from age 24-37 Malone was top 5 in PER and WIN Shares every single year, whiles being top 5 in the league in Points per game every single one of those years


Comparing the longevity of Karl Malone and Garnett

Seasons played

Malone-19
Garnett-21

Seasons played averaging 30+ Minutes

Malone-19
Garnett-15

Seasons averaging 20+ PPG

Malone-17
Garnett-9

Post Seasons averaging 20+ PPG

Malone-18
Garnett-5


All Nba 1st teams

Malone-11
Garnett-4

Seasons shooting 50%TS or more

Malone-19
Garnett-17

Seasons with an offensive rating of 100+ or more

Malone-18
Garnett-20

Seasons playing 80 or more games

Malone- 17
Garnett-8

Seasons missing 20+ games

Garnett-6
Malone-2

Seasons being Top 5 in WIN SHARES

Malone-13
Garnett-4

Seasons being Top 5 in PER

Malone-13
Garnett-5

Seasons being Top 5 in value over replacement player

Malone-10
Garnett-7

[Hornet Mania]

My votes go to:

12. Oscar Robertson
2nd vote: Julius Erving

Robertson was an all-in-one floor general that was well ahead of his time, the best "old-school" guard imo over West and Baylor.

Dr. J also was part of a generation of high-flyers (along with David Thompson and others) who truly made basketball a game played above the rim by elite athletes. I think his lack of elite ball handling has been overstated, he was fine for his era and had he been given the latitude to basically carry the ball like today's players get he could get by. An extremely graceful player, one of the most aesthetically-pleasing ever. If you go on Youtube and watch the old Lakers/Sixers Finals games (80 and 82 especially) Brent Musburger is practically beside himself with how awesome "The Doctor" was. His impact and popularity have been mostly forgotten, but he was a generational star in every way.

Players I'm considering in the next few spots: Karl Malone, Moses Malone, David Robinson, Dirk. I was tempted to throw an alternate vote to Mailman last thread but just couldn't do it in good conscience (I feel Oscar and Dr. J are simply better at their best), as soon as either Dr. J or Oscar goes in Malone will definitely be my alt vote.

[dontcalltimeout]

Some Regression Results...

Spoiler:

I finally got around to doing the regression I said I would in the first couple threads. What I did was very close to WOWYR (which you can read about here and here) but with four differences:

1. The lambda was chosen using generalized cross validation rather than k-fold cross validation.

2. By accident, I am missing any games from June, which is the finals and sometimes parts of the conference finals. I don't expect this to make a difference in the results, though I plan to eventually re-scrape for all the missing June data, and if there is a difference I'll be sure to make a post about it.

3. If you read Elgee's post on WOWYR, you would know it excludes players who do not meet certain minute thresholds. I do the same, though I made 25 minutes a strict requirement, rather than allowing for sub-cases where players playing fewer minutes can find a place in the regression.

4. Perhaps the biggest difference, I still use each player's MPG for the season, instead of MP each game, post-1984. I did this because I noticed the top players in Elgee's results are dominated by those who played after 1984. I became suspicious that the regression was more likely to underestimate player impact than overestimate, which we expect given the regularization, and thus by providing more granular data the regression can make better predictions which are not underestimations to the same degree.


Note: The results I arrived at are still dominated by players who played after 1984, but in comparison, their fitted values are not so clearly ahead of those for player's who played before 1984

One of the first things I did was run a regression on the whole data set from 1957 to 2017. This has advantages and disadvantages:

1. The advantage is the regression has plenty of data on all the players. Consistently, when doing the regression over smaller sets of years, the players at the top are those who played just one or two seasons. This also has consequences on players who played a reasonable number of seasons during the time frame, since their values are influenced by every one else's, and it's possible to find examples of including or excluding single seasons radically changing the fitted values for some players largely because what the regression thinks of the player's teammates changes.

2. The disadvantage is two fold. First, we are not isolating each player's prime. Jordan will get knocked for his Wizard years, Kareem and Duncan for their last couple seasons, Kobe for the start and end of his career, and so on. Second, the regression will assign only a single value for each player. Thus, players who played with very young or very old teammates, who became or were once great, will be knocked as well.

Here are the top-40, with players who played less than six full seasons not included.

[spoiler]Oscar Robertson: 8.17440313961
Kareem Abdul-Jabbar: 6.00160305044
Marc Gasol: 5.86949181662
John Stockton: 5.85137846504
Jerry West: 5.69099893136
Larry Bird: 5.42091725767
Michael Jordan: 5.17750091503
Magic Johnson: 5.0883282586
Vlade Divac: 5.03216295778
Bill Russell: 5.01597771069
Paul Millsap: 4.64913104721
LeBron James: 4.48963856742
Charles Barkley: 4.39735631558
Billy Cunningham: 4.28419502991
David Robinson: 4.18913711245
Wilt Chamberlain: 4.1675021247
Rashard Lewis: 4.1406827461
Paul Pierce: 4.13818986678
Cliff Levingston: 4.02393240086
Andre Iguodala: 3.99190507129
John Havlicek: 3.98714541027
Clyde Drexler: 3.97934152384
Dirk Nowitzki: 3.9751039827
Steve Nash: 3.95264201506
Hakeem Olajuwon: 3.91583926344
Paul Pressey: 3.84977757632
Dikembe Mutombo: 3.77061264129
Bob Lanier: 3.76607312567
Thaddeus Young: 3.72146426899
Ben Wallace: 3.60972293311
Chris Paul: 3.60561256364
Kevin Garnett: 3.53719182632
Nate McMillan: 3.39738875882
Tim Duncan: 3.35100925965
Doc Rivers: 3.19478703687
Willis Reed: 3.17223596463
Manu Ginobili: 3.16305929763
Bob Weiss: 3.14541255624
Bob Pettit: 3.08298603252

Some general comments:

- I won't dwell on it long, since topic has passed, but I feel obligated to mention it. For all the people saying that Kareem wasn't impactful, he's second all-time in this regression, and shows similarly well in regressions focused around specific player's primes. Aside from a couple ten-game samples there was never evidence to the contrary.

- Like WOWYR, we have a good number of unexpected names in the top-40, but otherwise we see all-time greats. The number of strange results that I got doesn't seem to be any more than what Elgee got which is reassuring.

- Oscar is far ahead of the pack. What's remarkable is he looks this great even if you don't include his Bucks years. My suspicion was that he was getting a massive boost from when he joined the Bucks and their SRS shot up but that wasn't the case. I'm surprised there hasn't been more support for him thus far.

A Problem...

So, while it's nice to have these results, there are problems. I'll walk through an illustrative example.

Let's look at Tim Duncan's results from 1998 to 2003. He's shows up as the very best with a fitted value of 5.78813166657 which isn't too surprising. Now, what about from 1998 to 2002? Duncan still show up well, his fitted value is 4.28670048637, but he's no longer at the very top. Was Duncan's 2003 year just so good it totally changes the results? Well, if we run the regression on just the 2003 season, Duncan looks like a 3.03965743306 (though results this season seemed deflated). Okay, let's extend the sample, and run it over 2002-2003. This time Duncan appears as a 2.77350967918. You should be thinking this is quite strange.

What's going on is Tim Duncan's fitted values are being greatly affected by his teammates. In the 1998 to 2002 regression, David Robinson is a 2.94718081154, while in the 1998 to 2003 regression Robinson is a 1.35647870915, and this is because he shows up as a -1.96055137035 in the 2003 season.

Five seasons isn't a large sample so we don't want to put too much stock into the results for Duncan and Robinson I've been discussing. However, it should be recognized that a player's fitted values can depend greatly on the regression's impression of other players, and in particular teammates. Even with a seemingly reasonable sample for some player of interest, it must be considered if the sample is large enough for the player's teammates, which makes running the regression on just a player's prime very difficult.

Now, with a larger sample regression, we don't expect this to be as much of a problem. After all, the regression will generally need to be more accurate in estimating the player who has ten years in the sample than the player with five years in order to minimize the loss function. However, this does prevent us from isolating a player's prime.

Providing an alternative that uses multiple seasons of data to yield estimates for smaller sets of seasons is something I will work on in the upcoming days. That, and measuring the error.

A Request...

I'm also going to make a request for help. I would like to get my hands on some RAPM (lineup) data to test the noise. The websites that had several years available for free are down as are the data sets which fpliii uploaded a year or so ago. I have a half-made scraper to get the RAPM data, but right now I want to focus on improving my WOWYR-type model, and I have a back log of research papers that I'm reading through. In other words, I'm lacking for time, and would really appreciate a helping hand. Even just a single year would be great if some one happens to have the data on hand.

These and Elgee's results indicate that Big O and The Logo are as great as their reputation suggests. Also, D-Rob and Stockton's strong performance is congruent with the limited RAPM results we have from them.

Some interesting results: Paul Millsap is high up there, though the regression doesn't include his inevitable decline. Iguodala too, though the way his role has been managed to provide maximum efficacy, I'm not surprised.
I am surprised at seeing Paul Pierce so high up.

Also, the one-way players: Charles Barkley, Nash, Deke, and Ben Wallace are up there. All players who were so good on one end of the floor that they were among the highest impact players in their primes.

[wojoaderge]

In 1977, Erving's statistics fell off due to poor roster construction.

I still don't like this. The roster was constructed around an All-NBA first teamer, and The Doctor was added to it the day before the season started. And still they finished with the best record in the EC for the two years.

[janmagn]

Vote: Karl Malone
2nd vote: George Mikan

Malone was great. He was physically just too good to be stopped with. He was elite finisher and rebounded very well. Good to great defender at the 4, even though stats might not show that.

Lähetetty minun LG-H440n laitteesta Tapatalkilla

[ardee]

I am voting Oscar here.

For starters, the guy's consistency is out of this world. From 1961-1968, he averaged no less than 28.3 ppg and 9.5 apg and no more than 31.4 ppg and 11.5 apg. And if you look at team correlation to those numbers, it is very high. He had them first in ORtg 6 out of 8 years, 2nd and 3rd in the other two. Great stuff. The rebounding is nice too, not that sure it would carry over without the pace, but I am pretty sure the rest of his game would translate very well to the modern game. I could see him as basically repeating CP3's '08 and '09 seasons for close to a decade, which is a big deal. Of course, era dominance is the main factor, but it does give me a few extra points if a guy's impact would be timeless.

His post-game is a very underrated part of his impact and why I think he could flourish on many teams. I could very easily see him playing like second threepeat Jordan on a team with shooters, backing guys down and drawing perimeter defenders in to kick the ball out, but also doing the PnR thing. His body contortion was really quite extraordinary, part of the reason he had the success he did against the Russell Celtics.

He's the clear choice for me here, he has the raw stats, advanced stats, high WOWY (if that matters to you), no real personality problems (I don't consider being a perfectionist a personality problem, otherwise you might as well take points away from Paul as well).

Vote: Oscar Robertson

Alt Vote: Jerry West. This is tentative, I would like to see someone break down Dirk vs West.

[Outside]

Vote: Jerry West
Alternate: Oscar Robertson

West was an elite player on both offense and defense, upped his game in the playoffs, was a great leader and teammate, combined athleticism and skill, had competitive fire. The only thing he didn't have was size, as in center-size. Dominated games from the guard position. Mr. Clutch. The Logo. An iconic figure in the sport.

Oscar used superior size, strength, and skill at the guard position to get where he wanted on the court. As I mentioned in another post, I downgrade him a bit because he could be a difficult teammate.

Below is my take on other candidates mentioned.

Dirk
I've got him at 33 (sorry)

Positives
Longevity
Scoring
Spacing
Efficiency
Low turnovers
Good playoff performer
Leadership -- not elite compared to many guys picked in prior threads, but still notable
Good teammate -- no drama, plays team ball

Negatives
Offensive diversity -- not a driver/slasher/finisher, not useful on the break other than as a three-point trailer. Despite his size, allergic to the paint (in his career, only 14.4% of his shots come inside 3 feet, only 8.4% from 3-10 feet).
Defense -- others have tried to make a case for him on D, but he is slow, not mobile, and not gifted with a defensive IQ or mindset. He did okay around the rim, but in space, he's just never been good. He was merely okay at best and has been a liability on D for most of his career. It's no coincidence that the Mavs' title team needed Tyson Chandler at center.
Weak rebounding for his size and position -- averaged only 7.8 for his career, never averaged double figures
Not a playmaker -- averaged only 2.5 assists for his career with a peak of only 3.5
Doesn't have the versatility in his game to be considered at this point

Mikan
I've got him at 27, but that's likely to change

Positives
Era dominance
Scoring
Rebounding

Negtives
No black players in his era
Exposed by athletic big men

Unknowns
Defense
Passing and playmaking

Julius Erving
I've got him at 21, but I could easily have him at 14. Players in this range are basically a 10-way tie for 14th.

Positives
Scoring
Offensive versatility
Longevity
One of the great finishers ever
Rebounding -- averaged 8.5 per game for his career from the SF position, somewhat inflated by ABA numbers
Not great, but good in misc categories -- assists, steals, blocks
Good playoff performer

Negatives
Mediocre efficiency
Long two shooter, not threes
Turnovers
Too much of an iso-ball guy

Moses Malone
I've got him at 15

Positives
Rebounding
Longevity
Energy and tenacity
Good scorer
Good playoff performer
Good defender

Negatives
Rebound number somewhat inflated by rebounding his own misses
Few seasons with deep playoff runs
Somewhat of a me-baller

Karl Malone
I've got him at 18

Positives
Longevity
Scoring
Offensive versatility -- inside/outside game, ran the floor, could drive, basically everything except three-pointers
Rebounding
Good (not great) in numerous areas -- playmaking, defense, turnovers, efficiency
Free throw shooting -- got to the line a lot, shot a good percentage
Good overall in the playoffs
Good teammate

Negatives
Poor 1997 finals
Bag of cheap tricks -- the Draymond Green high-kick specialist of his day

Other
Never won a title, shoulders blame for 1997 finals but played well in 1998 finals, was on teams with rosters that were good but not deep so I don't consider his lack of a title a negative

[Doctor MJ]

Hey, I believe someone posted something in the last thread that argued West's playoff performance was far better than Oscar's with statistics. Would you mind posting that again?

So it's out there, I'm leaning toward Oscar/Dirk right now, but West is a big wild card for me.

[mdonnelly1989]

VOTE #1: Oscar Robertson


This is where I feel like Big O was getting underrated because of the ERA he played in. I don't think someone like KG could ever anchor a team on total impact like Big O could.

KG was great, but he just didn't seem like the type that could carry a team to the promise land as the player who had by far the most impact. The 08 Celtics he was already passed his peak and was a part of a Fab 4 in Paul Pierce, Ray Ray & a Young but very good Rajon Rondo.

Oscar just seemed like in his time as quite possibly a top 10 player of all time esque. Do to the the dominance of Wilt and Bill Russell, they overshadowed his ability imo which why he doesn't get the recognition imo.


VOTE #2: Jerry West

Because I feel like Jerry West even though may have not been as a great as Big O in terms of overall impact and ability he was just a sliver underneath. And if you told me you thought West > Oscar I would have no problems with that as he was considered a MUCH better defensive player.

[BasketballFan7]

In 1977, Erving's statistics fell off due to poor roster construction.

I still don't like this. The roster was constructed around an All-NBA first teamer, and The Doctor was added to it the day before the season started. And still they finished with the best record in the EC for the two years.

The team had a lot of talent. That doesn't speak to its roster construction, the roster's cohesiveness. McGinnis, Collins, Free, and Erving already make for too many mouths to feed offensively. There is no fit there. Erving being a late addition gives the issue reasonable cause but does not remedy it. Fred Carter was another high volume shooter who ended up missing nearly the entire season.

The Big 3 Heat had poor roster fit early on. That doesn't mean that their management did a poor job or that they had a bad team.

[trex_8063]

Spotty internet here. Have been reading some, made much of a long post only to have it vanish when my connection dropped

I'll be brief......
1st vote: Karl Malone
2nd vote: Dirk Nowitzi

Reasons to follow before I lose connection