colts18 wrote:I'm still questioning whether Wilt had a great impact.
In 1968 the Lakers had a 4.99 SRS without Wilt.
In 1969, the Lakers, a team in desperate need of Center help, acquired Wilt (winner of the 3 previous MVP) and had Jerry West play 10 more games than the previous season. Did the Lakers get better? No. In fact their SRS declined to 3.84 SRS. That's an indictment on Wilt. Could you imagine if the Thunder acquired Peak Shaq (2001) and had Westbrook play more games. Would they become a worse team? Absolutely not.
You don't understand the circumstances of the time, obviously.
Wilt was used perfectly in Philadelphia, alternately as a low-post scorer and facilitator. In any case, the ball went into him every time.
Van Brenda Kloff, idiot that he was, moved Wilt out to the high post to accomodate an aging Baylor. For whatever reason, he acquired the best player in the league and decided to take him out of his best spot. This caused a season long feud between Wilt and VBK that caused difficulties within the team, through no real fault of Wilt's.
Now tell me, a guy like Shaq, with no range at all: if you put him in the high post, would his impact be worsened? Hell yes it would.
Your post smacks of a guy who looks at BBRef without considering the circumstances at all.
What's your basis for that? I wouldn't call VBK an idiot -- it's a lot like calling Mike D'Antoni an idiot. VBK implemented Princeton-style offensive sets that were ridiculously successful in 1968. We're talking record-setting stuff. VBK clashing with Wilt and how Wilt was supposed to fit into that "system" does not mean he was an idiot, nor that he was trying to accommodate for Baylor.
Good call. 69 was a pretty clear case of a team already great with a state of the art offense. If Wilt had come in and just asked himself "where can I help" there's every reason to believe the team becomes tremendous.
Instead after Basically forcing his way out of a great situation in Philly so he could live in LA, he proceeded to get but thrust when the Lakers didn't tear down what they were doing for him.
To look at that situation and assume stupidity on the part of the coach is exactly what I mean when I talk of people not putting themselves into the point in history they are trying to understand and as a result falling prey to the precise kind of naively they allege of the historical figures.
General rule if thumb: assume humans are humans same as you and only use the diagnosis of brilliance and idiocy as a last resort.
Tossup for me between Magic/Kobe/Duncan, but the Duncan arguments have consisted mostly about his "weak support", and Kobe clearly has no votes here, so I'm breaking the tie and going back to my pick from last time of Magic.
My 5 left at each postion:
PG - Magic: 11 off / 4 def SG - Kobe: 10 off / 5-6 def SF - Lebron: 9-10 off / 6 def PF - Duncan: 6 off / 8-9 def C - Shaq: 10 off / 5-6 def(7-8 def for 00/01) Wildcard - Wilt: 10 off / 4-8 def(big evolution over the years)
Lebron doesn't have the career value yet to pass any of these guys, but I think by the next list he'll be much higher. Lebron's boxscore to team impact ratio seems off too. he's had 7 contending teams since his prime started in 2009, yet got just 2 rings in that span(good, but not better than the others listed), and his teams lost to inferior opponents 3 out of the 7 years.
Shaq/Wilt are pretty even, but I have them both below Magic, Kobe, Duncan right now. My big issue with them is their on court focus & defense as bigs. They're easily the two most dominant player ever within 6 feet, however they were limited to the paint area for offensive impact, not to mention the FT troubles which caused alot of 1 and outs. Dipper had a good post analyzing Wilt in a 1973 game. viewtopic.php?f=64&t=1247724 His style was eeriely similiar to Shaq on both off/def. Wilt's on court production trumps Shaq's, but how much of that is due to era? More questions than answers for these 2 at this point.
Kobe/Duncan are extremely close, but I give Kobe the edge considering he got the best of Duncan during their primes. Both players met 6 times in the playoffs, and in on court impact Kobe had the edge 4 out 6 of the meetings. Duncan's defensive impact was great, but no matter how good an individual defender is, he can't guard more than 1 player, hence Kobe dropping 45 against the Twin Towers, and there's no answer Duncan could give. Reminds me of the Ewing/MJ battles in the 90's. I also favor Kobe's prime which was better/longer, imo.
But I'm going with Magic since he's the most viable nominee, and virtually tied with Kobe in my view on impact.
-GOAT offensive player. Had to put him at 11 to show proper context -GOAT passer -GOAT mismatch -GOAT level leadership -underrated scorer -solid rebounder
Clearly his BIG flaw is on defense, no way around that. But again, Great individual offense is unmatched impact wise. Magic didn't just get himself producing, he got the whole squad working like a well-oiled machine.
Why do you have KG ranked higher than Shaq? Shaq was consistently better in RAPM stats and was a much better playoff performer.
I think some people just like to go against the norm.
So first off, just stop it. A post that says nothing about basketball and just disparages the poster is not tolerated here.
Second though I'll address your feelings just pretending you expressed them with more tact.
There's an ongoing meme on this board where people say "only on realgm" as a way of saying how far afield some opinions are relative to norms. What's odd about it to me is the apparent contradiction. If people are are so misguided, why spend time here?
The answer, of course, is that people come here because they recognize this is one of, or perhaps, THE, place to be for really good NBA discussion based on the obvious depth of the conversation...but then when they are brought face to face with where that leads to in terms of developed opinions they get cognitive dissonance.
I don't want to appeal to authority here. No one should agree with something simply because I or some other person says it here. I will say though that it would be foolish for anyone to brush these afield opinions off lightly.
A good rule of thumb: Assume on here that any conventional opinion you came here with is completely understood by the regulars and that if they disagree, it's because something powerful changed their mind. If you don't understand what that was, then learn a thing.
MacGill wrote:I am hoping Reg or someone could paste some great Magic posts of the past.
I can post the synopsis I have of Magic from my notes then if you'd like.
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
Had a chance to estimate off/def splits (had to approximate FGA...I calculated for teams with Wilt by dividing FG during the sample by the team FG% including Wilt but excluding Neumann/Dierking, and calculated for teams without Wilt by dividing FG during the sample by the team FG% including Neumann/Dierking and excluding Wilt):
The estimates aren't perfect, but they give a ballpark idea. Wilt leaving SF made the defense much better, but the offense much worse. Wilt arriving in Philly made the defense worse, but the offense better.
This is interesting for a couple of reasons:
1) It supports the idea that 65 in general wasn't a good defensive player, regardless of where he played. This was true in SF, in Philly during the season, and in the playoffs. This could have to do with the appendix thing, but was it really bothering him the whole year?
2) It's interesting that 65, when looked at in this light, resembles 62 and 63 more than 64. It makes one wonder how much of an effect Hannum truly had on Wilt in 64. The weird thing is, the next year in Philly, Wilt produced similarly in the regular season offensively, but had a good defensive season (though not so much in the playoffs).
Not enough to sway me away from Wilt since I already had an idea of what the numbers would look like, but if Hannum in 64 didn't leave a very long-lasting impression at all on Wilt, then it adds more questions IMO about how Wilt would perform in general without all-time great coaches (such as Hannum and Sharman).
Now that's the difference between first and last place.
drza wrote:In order, no he wasn't, and that's worth discussion.
I've watched how you use RAPM over time. On the one hand I applaud it, because I remember your arguments when you first started posting here and they've gotten more sophisticated and supported, which is a great thing. On the other hand, I've also noticed that you tend to be very selective with using different types of RAPM to better fit your case without being clear on that. You've already done it in this thread, pulling out XRAPM data and calling it RAPM (when you KNOW it isn't the same thing). This can be dangerous, especially in this environment where any type of "advanced" statistical analysis is held in such suspicion. Selective/preferential application of stats as you do can start to smack of those that believe in the "lies, damn lies and statistics meme".
How that relates, here. The "state of the art" (as it were) of non-boxscore based RAPM, is the prior informed approach. As I'm sure you're aware, KG doesn't take a back seat to anyone in that approach, including Shaq.
How is what I'm doing is selective yet you based your conclusion on a spreadsheet (Dr. MJ's) that is very selective. Are you shocked that KG would come out ahead in a ranking that either doesn't include or is incomplete for Duncan's 3 peak years (01-03)? Is it a surprise when KG comes ahead of Shaq in a spreadsheet that doesn't include arguably Shaq's best season?
btw, xRAPM is a very reliable stat. In fact more reliable and stable than RAPM. I don't get why this board hates the stat.
An Unbiased Fan wrote: Lebron's boxscore to team impact ratio seems off too. he's had 7 contending teams since his prime started in 2009, yet got just 2 rings in that span(good, but not better than the others listed), and his teams lost to inferior opponents 3 out of the 7 years.
I am sorry but saying he has had 7 contending teams since 09 is not true at all.
JordansBulls wrote:The Warriors are basically a good college team until they meet a team with bigs in the NBA.
An Unbiased Fan wrote:Lebron doesn't have the career value yet to pass any of these guys, but I think by the next list he'll be much higher. Lebron's boxscore to team impact ratio seems off too. he's had 7 contending teams since his prime started in 2009, yet got just 2 rings in that span(good, but not better than the others listed), and his teams lost to inferior opponents 3 out of the 7 years.
Where do you get 7 contending teams from? What are the 3 inferior opponents that LeBron lost too?
I've had Hakeem, Duncan and KG ahead of Magic and Bird for a few years, I feel the defensive impact while still being awesome offensively is too much to make up for and that the media preferred Magic/Bird peak in their era because of a bias towards offense and winning (I'll get to him later but Bird is who I'm fading most compared to previous editions. My position on him for a few years has been he has a top 10 peak, but not by very much, like 9th or 10th or something. When added to some longevity issues and playoff shooting % concerns in some years, I'm looking at a mid teens rank for him right now. I feel pretty strongly that Kobe, Duncan, KG, Lebron should all be ahead of him when considering skillset and longevity combined)
Why Hakeem over Duncan and KG: This one is really close since I prefer Duncan and KG's longevity and leadership. But... I feel Hakeem is the greatest of the three offensively at his peak and his "playoff performance bump" legit matters.
Why Hakeem over Shaq and Wilt: Shaq and Wilt are better offensive forces and probably even more talented players. Shaq's health, pnr defense and free throw % are all little hairs that bug me. Wilt has some playoff bruises. Both guys have higher peaks than even Hakeem (not by much) but Hakeem's value is probably more consistent and reliable. To be honest I'm going to read through this thread to make sure I'm right picking Hakeem over Shaq/Wilt. Or frankly just Shaq for now since I'm voting Shaq over Wilt anyways since they're on the opposite ends of knowing what they are
Doctor MJ wrote: Colts please stop referring to this without some acknowledgement of the discrepancy between studies.
You're using non prior informed RAPM and both pure APM and prior informed RAPM give Garnett the edge over Duncan and make Garnett vs Shaq look quite close.
How can you attack my post without acknowledging that your spreadsheet uses a NPI RAPM that has only 30% of the season completed? Shouldn't it make you curious why Duncan is ahead of KG in NPI RAPM every year from 01-04 yet KG has a higher prior informed RAPM that season? J.E. has acknowledged that the early seasons of RAPM (02-06) are incomplete yet you take it as the gospel. It should make you reconsider your stance when I point out discrepancies like that. It would make no sense that KG is ahead of Duncan in 2004 if Duncan has him beat every year prior. That only happens with incomplete data.
Doctor MJ wrote: Colts please stop referring to this without some acknowledgement of the discrepancy between studies.
You're using non prior informed RAPM and both pure APM and prior informed RAPM give Garnett the edge over Duncan and make Garnett vs Shaq look quite close.
How can you attack my post without acknowledging that your spreadsheet uses a NPI RAPM that has only 30% of the season completed? Shouldn't it make you curious why Duncan is ahead of KG in NPI RAPM every year from 01-04 yet KG has a higher prior informed RAPM that season? J.E. has acknowledged that the early seasons of RAPM (02-06) are incomplete yet you take it as the gospel. It should make you reconsider your stance when I point out discrepancies like that. It would make no sense that KG is ahead of Duncan in 2004 if Duncan has him beat every year prior. That only happens with incomplete data.
Two things:
1) Has J.E. acknowledged that 03-06 are incomplete? I've heard 02, but not the other seasons. I'd appreciate if you could link me to what he's said on this, thanks.
2) Not necessarily, as far as I can tell. The effect of adding a prior is that it increases the sample possessions for each player, and, if they play for different teams or in different lineups, it'll help eliminate collinearity by giving us more possessions without common pairs in lineups. So it's possible that NPI would report that one guy consistently performed in his role better than another player in a similar role, due to lack of sufficient possessional data for their teammates.
Now that's the difference between first and last place.
1) Has J.E. acknowledged that 03-06 are incomplete? I've heard 02, but not the other seasons. I'd appreciate if you could link me to what he's said on this, thanks.
Here it is.
Can you explain the prior-informed RAPM and how it works? In the non-informed data, Duncan leads from 01-07 every single year, yet KG's 2003 and 2004 prior-informed is higher than Duncan's. If you are using prior data, wouldn't Duncan be higher since he is higher in every single season?
"There's a discrepancy here because the data I used for computing informed RAPM was incomplete. For most of those early years it was missing >15%"
Doctor MJ wrote:A good rule of thumb: Assume on here that any conventional opinion you came here with is completely understood by the regulars and that if they disagree, it's because something powerful changed their mind. If you don't understand what that was, then learn a thing.
Very good rule. These projects have become a bit of a futile effort, I was really hoping we'd see a bit more evolution in the discussion this go around. It's still people posting the same generic conclusions that are short on details, and have been floating around for years and years. Given how much discussion there has been in the past, there is really no excuse to be posting so much of this stuff that's already been addressed. Over time, we should be drawing deeper and deeper distinctions to get a better understanding of a player, but instead it's the same superficial nonsense. I mean, a few pages back there's a quoted post about Wilt in elimination games that I made probably like 6 years ago, and half the pro-Wilt posts here probably accurately represent my opinion of him 6-7 years ago, go back long enough, you might even find a post about why I thought Hakeem was only a borderline top 10 player, but something happened. I̶ ̶l̶i̶k̶e̶ ̶t̶o̶ ̶t̶h̶i̶n̶k̶, I KNOW it was because I continued to evolve and acquire a richer, more detailed understanding of the careers of most of these players (as well as basketball in general) over time and that my previous opinions were due to a lot of gaps in my knowledge (ie. ignorance). I don’t care if someone agrees with my view point or how I choose to rate players, in fact, I like it better when someone brings a new insight that changes or refines my opinion but unfortunately, we're presented with the same stale material and arguments that a) everybody already has or should have a good understanding of and b) doesn't actually address the concerns that people have pointed out over time (which are either totally ignored or misunderstood). It doesn't matter if Wilt goes 4th or 12th, but the discussion surrounding it should be a lot better than it is. I'm assuming a few other posters are also glossing over the pages and rolling their eyes at the reasons being given, it's not some kind of a hipster arrogance, it's because most people are still anchored to opinions that the "hipsters" they are arguing with already had and evolved from over the years in light of acquiring more information, more research and engaging in many discussions. The result is that discussions stalls as there's pages and pages of shallow analysis and things that simply don't matter. 5 years ago it might have been fine, informative even, but now, it's just trite and reeks of ignorance.
1) Has J.E. acknowledged that 03-06 are incomplete? I've heard 02, but not the other seasons. I'd appreciate if you could link me to what he's said on this, thanks.
Here it is.
Can you explain the prior-informed RAPM and how it works? In the non-informed data, Duncan leads from 01-07 every single year, yet KG's 2003 and 2004 prior-informed is higher than Duncan's. If you are using prior data, wouldn't Duncan be higher since he is higher in every single season?
"There's a discrepancy here because the data I used for computing informed RAPM was incomplete. For most of those early years it was missing >15%"
TrueLAfan wrote:I give Shaq a bonus that not everyone gives him—but it’s a double edged sword. I don’t necessarily think Shaq peaked in 1999-2002. I think he was close enough to his peak so that you can say his peak was really 10 years long--from 1994 to 2003. That gives Shaq a long peak period for a superelite player.
The two problems this brings up for Shaq, though, are:
1) Conditioning and missed time. Shaq missed 127 games between 1993-4 and 2002-3. That’s over 16% of his team’s games. And he wasn’t always in the best of shape when he was on the court. Supporters point out that he didn’t miss playoff games. I know, but the impact from losing your star player for 13 or 14 games every year and not having him at 100% effectiveness for many others for largely avoidable reasons takes a toll on a team in measurable ways (Season records and HCA) and less obvious ways (What does it do for team morale to have your main player distracted by food, movies, and being a rap mogul?). 2) When he was at his long peak and other Top 20 Cs were at their long peak, there was no consensus that Shaq was the best C. It’s no insult to be at the level of Hakeem and Robinson, obviously. But Shaq missed a lot more time than those players did and for worse reasons. I can see the argument for Shaq’s peak being a little better. But, for a substantial chunk of his top years, he didn’t do as much with it.
Re: point #1) I specifically brought up missed games as a concern for me, but it's something Shaq supporters haven't addressed, as they're trying to sway people to vote for their candidate.
Spoiler:
ThaRegul8r wrote:Shaq’s longevity is often cited by Shaq supporters, but from 1992-93 (his rookie season) to 2005-06 (his last year selected to the All-NBA First Team), Shaq played 81, 81, 79, 54, 51, 60, 49 (50-game lockout season), 79, 74, 67, 67, 67, 73, and 59 games. Subtracting the lockout, that’s an average of 68.6 games per 82-game season. During discussion on Bill Walton during the Highest Peaks project, one poster said:
[spoiler]
An Unbiased Fan wrote:If durability counts for anything, it’s a bit hard to look at a guy who gives me only 65 games. That’s a bit fragile, is it not.
Yet Shaq played a mere four more games than Bill Walton played in ’1976-77 on average for his first 13 full seasons. I've wondered how one can reconcile that Walton is "fragile" for only playing 65 games, yet Shaq averages 69 for 13 years, but that isn't a problem. Wilt, on the other hand, was an ironman. He had only one major injury in his career, when he missed 70 games in 1969-70. Outside of that season, he averaged 79.5 games a season. Phil Jackson wrote of Shaq:
Spoiler:
Phil Jackson wrote:He’s often maligned for his lack of durability, his unwillingness to play with severe physical discomfort, yet the critics have no clue to what he must regularly overcome to compete at this level. Nobody can begin to understand what it must feel like to haul a 340-pound body around, stopping and starting, stopping and starting. There are players in this league who can’t play with a hangnail, but Shaquille O’Neal is definitely not one of them. Besides, there’s a discernible advantage to his sitting out fifteen or twenty games a year. The reduced wear and tear will help to preserve his knees for May and June when we will need him for forty-plus minutes a night. I see an analogy between Shaq and a 1972 220SL Mercedes. In 2004 one is not going to race this car at one hundred miles per hour down Sunset Boulevard because it would risk damaging the vehicle. How many times do you expect Shaq to run up and down the floor before something goes wrong? Shaq plays with an orthopedic device in each shoe that weighs about three pounds, forcing him to wear a size twenty-five instead of twenty-two. When he jumps or runs, he pushes off the outside of his foot instead of his big toe.
Now, that same size which contributed to his "unstoppability" also hampered his durability. Now, people can say they're okay with that so long as Shaq is ready to perform in the postseason, but I find it disingenuous when Shaq's longevity is cited without mention of his durability, when another player is "fragile."
[/spoiler]
Re: point #2) From my notes on my entry on Shaq, Shaq didn’t come of age until all the other great centers either retired or passed their prime. He wasn’t First Team All-NBA until the 1997-98 season, his sixth in the league, when Hakeem Olajuwon was 35, David Robinson was 32, and Patrick Ewing was 35. In his rookie season, Olajuwon (26.1 ppg, 13.0 rpg, 3.5 apg, 4.17 bpg, 1.83 spg, .529 FG%, DPOY), Ewing (24.2 ppg , 12.1 rpg, 1.9 apg, 1.99 bpg, .503 FG%) and Robinson (23.4 ppg, 11.7 rpg, 3,7 apg, 3.22 bpg, 1.55 spg, .501 FG%) were the All-NBA First, Second and Third Teamers. In ’93-94 it was Olajuwon (27.3 ppg, 11.9 rpg, 3.6 apg, 3.71 bpg, 1.6 spg, .528 FG%, MVP, DPOY) and Robinson (29.8 ppg, 10.7 rpg, 4.8 apg, 3.31 bpg, 1.74 spg, .507 FG%), with Shaq making Third Team. In ’94-95 he was Second Team behind David Robinson (27.6 ppg, 10.8 rpg, 2.9 apg, 3.23 bpg, 1.64 spg, .530 FG%, MVP). In ’95-96 he was Third Team behind Robinson (25.0 ppg, 12.2 rpg, 3.0 apg, 3.30 bpg, 1.35 spg, .516 FG%) and Olajuwon (26.9 ppg, 10.9 rpg, 3.6 apg, 2.88 bpg, .514 FG%). In ’96-97 he was Third Team behind Olajuwon (23.2 ppg, 9.2 rpg, 3.0 apg, 2.22 bpg, 1.5 spg, .510 FG%) and Ewing (22.4 ppg, 10.7 rpg, 2.0 apg, 2.42 bpg, .488 FG%). O’Neal was All-NBA First Team in ’97-98, Second Team behind Alonzo Mourning (20.1 ppg, 11.0 rpg, 1.6 apg, 3.91 bpg, .511 FG%, DPOY) in ’99, then was First Team from ’99-2000 to 2005-06. As you said, there was no consensus he was the best center, and no one was calling him the "MDE."
Then, as I brought up, by the time Shaq reached his absolute peak in 1999-2000, when the Lakers’ season began on Nov. 2, O’Neal was 27 years, 241 days old. Olajuwon was 36 years, 285 days old; Robinson was 34 years, 88 days old; Ewing was 37 years, 89 days old. I posted articles of the time commenting on the lack of competition for Shaq now that he was at his best. There was no answer for this because there is none. The assumption made is that if Shaq "handled himself well" against the elite centers pre-peak, then it would be reasonable to assume he would do just fine against elite competition in his peak. Though that doesn’t change the fact that that competition wasn’t there when Shaq reached his zenith, as was commented on at the time. Phil Jackson before the start of the 2000-01 season:
Spoiler:
Phil Jackson wrote:Alonzo is ill, Mutombo has some physical problems, too. Ewing is in Seattle and Smits is retired. That’s four major centers stepping away from the game.
Since I'm factoring in everything, not just the parts people want to talk about, that's on the other side of the ledger. And to top it all off, instead of exploiting the lack of competition, he failed to take care of himself, and you had stuff like this:
Spoiler:
Off the Glass Why isn't Shaq the most dominant O'Neal in the league? Posted: Thursday March 11, 2004 8:19PM; Updated: Thursday March 11, 2004 8:19PM
By Paul Forrester
A few years back, OTG engaged in a vigorous debate with SI.com's own Jennifer Cooper about the talent of a certain NBA giant named Shaquille O'Neal. Having watched the Daddy plow over helplessly outsized opponents in her SEC stomping grounds, Jennifer was of the opinion that O'Neal was little more than a bully, using his tremendous size advantage to seemingly score at will and control the paint.
Hogwash, said a young(er) OTG; no matter the size, a person must have some level of talent to understand how to use his bulk effectively. The league is littered with the memories of Mark Eaton or Gheorge Muresan, giants who played nice complementary roles but never displayed the game-changing ability of Shaq.
On the debate raged, over the course of days, weeks, months, years, neither one of us giving into the other's argument.
Ms. Cooper, can I change my answer?
Maybe change is too strong a word. Amend might be more appropriate. As dominant as everyone tells us Shaq is, there is something missing, that sense of wow. I'm not looking for a skywalking dunk or a behind-the-back pass. I'm talking about putting up the kind of statistics that create headlines: 40 points for five games in a row or 23-rebound nights, triple doubles in points, rebounds and blocks, things that Shaq just doesn't do.
True, Shaq did do those sorts of things when he came into the league, but OTG has always had the sense that the Daddy didn't have the drive to excel in hoops to his full, monstrous potential.
As the years have added up, Shaq has increasingly picked his spots as to when he wants to dominate. Recall last year when Shaq didn't opt for toe surgery until late in the offseason, guaranteeing him a late start to the regular season.
Or look at his numbers this year: 22 ppg (a career low) and 10.8 rpg. Heck, this season, on a pound-for-pound basis, Shaq isn't even the best O'Neal in the NBA.
Indiana's Jermaine has averaged 20.6 ppg, 10.3 rpg and 2.6 bpg carrying 98 fewer pounds and measuring three inches shorter. Sure, Shaq is hitting nearly 60 percent of his shots, compared to Jermaine's 43 percent, but shouldn't the league's most dominant player, a two-time Finals MVP, own more than one statistical category?
Heck, yes, he should, especially in a season in which injuries to Karl Malone and Kobe Bryant have left the Daddy with no more offensive options than he has had in the past. I understand Shaq tries to preserve himself for the postseason, but he plays on cruise control far too often, a habit that will leave the Lakers without home court in the playoffs' later rounds.
While O'Neal has always been a sometimes indifferent defender, his lack of focus has drifted toward his offensive game lately. How does someone of Shaq's size fail to score 20 points or pull down more than five boards against Jahidi White of the Suns? Samuel Dalembert may show signs of being a solid center for Philly but should a rookie be able to restrain Shaq to 17 points, eight rebounds and a single block? Do you catch my drift? Seventeen, eight and one are fine numbers but the Daddy is capable of so much more, it's a pity he doesn't do more.
By all accounts he cannot be guarded when he wants to get to the hoop. Far too many times, though, he is, and guarded out of relevance for long stretches of games. That's the reason I can write a column like this and be reasonably certain that children will not throw eggs at stately OTG Manor. That's the reason the Lakers need a one-shouldered Kobe Bryant to bail them out in the fourth quarter. And, finally, it's the reason I suspect I may finally have lost that old debate.
And the bolded illustrates what you were talking about regarding injuries, and how season records if Shaq, as Phil Jackson said, can be counted on to miss 15-20 games a year, that will affect your record, which will affect what seed you get in the playoffs. That could be the difference between a favorable matchup and one that is disadvantageous to you. With all the talk about HCA, still is significant, and will be factored in. If you don't have HCA because Shaq missed too many games, then it's affecting the team's chances to win. Another center who's going to be on the court longer for you can help you get a higher seed for the postseason.
The availability heuristic is evident with Shaq, as everyone remembers him dominating, but no one remembers him as less than that anymore. When it comes to Shaq, a bunch of numbers are thrown around, but no one has yet to alleviate any of the concerns I have about him.
Now, since it isn't just Shaq, and I'm not just picking on him...
TrueLAfan wrote:The discussion and selection between Shaq and Hakeem is interesting. I think Hakeem has become a little bit overrated; there’s revisionist history going on. I read a lot about how his peak wasn’t in the mid-90s, but started far earlier in his career and lasted longer. This is not what people thought at the time. Between 1986 and 1990, Hakeem was a consensus Top 5-6 player. He got slightly over .20 in MVP win shares in two of those years. That is not commensurate with a Top 3 player. Hakeem was never considered to be as valuable as Magic, Michael, or Larry. He was in a group with Dominique and Barkley and Malone in that period (those other players actually had more MVP win shares than Hakeem in that span.)
I addressed this as well. As I've said before, Hakeem was my favorite of the 90's crop of centers at the time. But revisionism has occurred with him. His two rings has caused people to elevate his entire career. I'll quote myself:
Spoiler:
ThaRegul8r wrote:
The Infamous1 wrote:
ThaRegul8r wrote: The bolded is a concern of mine.
Some people in previous discussions brought up Wilt's All-NBA lead over Russell. Being consistent and using the same criteria, from 1991 (his second year in the league) to 1996, David Robinson was a four-time All-NBA First Teamer (’91, ’92, ’95, ’96) to Olajuwon’s two (’93, ’94). Then in 1996-97, Robinson missed 74 games with a broken foot. So up until the injury—again, using previously used criteria, Robinson was seen as > Hakeem. Looking at MVP Award Shares, from 1991 to 1996 Robinson had 2.97 MVP Award Shares to Hakeem’s 1.895. Whereas in the Wilt/Russell example, in the years their careers overlapped, the MVP Award shares were 3.771 Wilt/3.883 Russell. So it was a lot closer between those two than between Robinson and Hakeem.
But then '95 happened, and changed everything.
Not just drob, at many points during his career guys like Ewing,Barkley, a Malone were all considered better or at worst on par with Hakeem.
It wasn't until the 94 and 95 playoffs that he seperated himself from them historically
Exactly.
It seems some revisionism has occurred with time, and those rings have caused people to see him differently from how he actually was. That's why—as I said earlier—I want to cut through the hyperbole in order to see where to properly place him.
Now, if someone wants to separate rankings into Peak and Career like Bill James does for baseball, then fine. But Hakeem's peak has elevated his entire career to where perhaps it shouldn't be.
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
Just a note, I posted playoff off/def splits for Wilt earlier, and found a minor error in my calculations (I included OTs in calculating pace, but not in adjusting PPG before dividing by pace. Here are the corrected numbers: and the originals, for comparison's sake: Apologies for the mistake. Doesn't change all that much, but I feel more secure about Wilt's defense in 65 and 66 since it wasn't too bad. 68 and 71 also look better on that end. All four years (65, 66, 68, 71) look slightly worse offensively.
Now that's the difference between first and last place.
ardee wrote: This post is intellectually dishonest.
To make Gola, Arizin and Rodgers out to be really valuable teammates based on their HOF status is dishonest. Gola and Arizin were near the end of their primes when they played with Wilt in the early 60s. Rodgers was an awful offensive player. Couldn't shoot or really score in anyway, never cracked 40% from the field.
And Thurmond was in his rookie and sophomore year as Wilt's backup when they played together.
The best teammate Wilt had in his prime was Hal Greer, who is probably about the same level as Sampson and definitely worse than Drexler.
Sent from my GT-I9300 using RealGM Forums mobile app
Particularly so because it it compares accolades without the context of league size (and particularly All-Stars whereby there was a maximum of three all-stars per team which for much of the 60s meant you were depending on your conference guaranteed 3 players, or once the Chicago-Baltimore franchise arrived in the east merely likely to get 2 or 3) whilst ignoring the more representative all-NBA teams.
I wouldn't deny Hakeem didn't have the supporting casts to succeed in the late 80s, but to imply Wilt did based on HOF (Rodgers long, long after the fact, Gola based on HS/college, Arizin based primarily on his 50s, though still effective in the early 60s) is misguided.
Based on the evidence I've seen, and keep finding, these guys were good players.
I wanted to address these arguments. The more I read about these guys, relative to competition they were quality players with plenty of accomplishments, not just All Star or HOF, but All NBA selections, Top 50 player selections etc.
Gola played with Wilt from age 27-29, and was 2nd team all NBA the year before. He wasn't even in his thirties yet. He was also one of the greatest college players of all time.
La Salle
Gola was one of the most talented collegiate athletes in Philadelphia sports history. He came to national attention while playing for the hometown La Salle University Explorers men's basketball team.
Gola starred as a college freshman and led La Salle to the 1952 N.I.T. championship.[2] Gola paced the Explorers to the NCAA basketball championship in 1954 and was named Tournament MVP. That same season he was selected as National Player of the Year. As a senior, Gola helped La Salle finish as the runner-up in the 1955 NCAA Tournament. He averaged 20.9 points and 19.0 rebounds during 115 games and holds the NCAA record with 2,201 career rebounds. At 6'6" (198 cm), Gola was clearly a forward who could shoot/score, rebound and defend, but he also had the ballhandling (dribbling, passing) skills of a guard, and with his shooting range and All-Pro defensive skills, could play just as well in the backcourt. He was inducted into the La Salle Hall of Athletes in 1961 and the Big 5 Hall of Fame in 1986. In 1977, Tom Gola was inducted into the National Polish American Sports Hall of Fame.[3] Gola was listed on "ESPN's Countdown to the Greatest" College basketball players as #17.
NBA
After a phenomenal college career, Gola turned pro with the Philadelphia Warriors as a territorial draft pick. He teamed with All-Pros Paul Arizin and Neil Johnston to lead the Warriors to an NBA championship in 1956. He gained praise for concentrating on defense, passing and rebounding and allowing the other two to be the chief scorers during these years.
In 1959, Johnston temporarily retired due to a knee injury and the Warriors added seven-foot superstar Wilt Chamberlain. Again sacrificing himself for his team, Gola helped the Warriors back to the NBA Finals, but they could not beat the star-studded Boston Celtics during his seasons in the early 1960s even with Chamberlain.
Paul Arizin was 31- 33 and All NBA 2nd team Wilt's first year (before that he was All NBA first team for three years).
Professional career After being selected by the Warriors with their first pick in the 1950 NBA Draft, Arizin averaged 17.2 points per game in his rookie season and was named NBA Rookie of the Year — a designation not currently sanctioned by the NBA for the 1950-51 season. He became one of the greatest NBA players of the 1950s, leading the NBA in scoring during the 1951–52 and 1956–57 seasons and leading the league in field goal percentage in 1951-52. Arizin sat out the 1952–53 and 1953–54 NBA seasons due to military service in the Marines during the Korean War.[4]
Arizin became famous for his line-drive jump shots, and teamed with center Neil Johnston to form the best offensive one-two punch in the NBA at the time, leading the Warriors to the 1956 NBA title. He also played with scoring star Joe Fulks early in his career, and with Philadelphia legends Tom Gola and Wilt Chamberlain toward the end of his career in the early 1960s. Arizin chose to retire from the NBA rather than move with the Warriors to San Francisco. At the time of his retirement, no player had retired from the game with a higher scoring average (21.9 points per game) in his final season. This record would stand until Bob Pettit's retirement in 1965 following a season in which he averaged 22.5 PPG.
Arizin played in a total of 10 NBA All-Star Games (he was the 1952 NBA All-Star Game MVP) and was named to the All-NBA First-Team in 1952, 1956, and 1957.
After retiring from the NBA, Arizin played for three seasons with the Camden Bullets of the Eastern Professional Basketball League, who won the 1964 title. Averaging over 20 points per game each season, Arizin was named the EBL MVP in 1963, was named to the EBL All-Star First Team in 1963 and 1964, and was named to the EBL All-Star Second Team in 1965.[5]
Arizin was named to the NBA 25th Anniversary Team in 1971. He was inducted into the Naismith Memorial Basketball Hall of Fame in 1978,[6] and was selected to the 50 Greatest Players in NBA History in 1996.
He may have been in his thirties but he was still a good enough talent to make the 50 Greatest Players List and All NBA several times before. I didn't grow up in the era, but I have a hard time seeing him as a quota all star with those other accomplishments relative to his competition in that league.
Guy Rodgers was age 24 - 29 (prime imo) and an excellent play-maker for Wilt. His strength was passing not scoring:
Guy William Rodgers (September 1, 1935 – February 19, 2001) was an American professional basketball player born in Philadelphia, Pennsylvania. He spent twelve years (1958–1970) in the NBA, and was one of the league's best playmakers in the early to mid-1960s. Rodgers led the NBA in assists twice, and placed second six times. [1]
Rodgers played alongside the great Wilt Chamberlain from 1959 through 1964, and during Chamberlain's famous 100-point game, he led the way with 20 assists.[2] In the 1962–63 season, Rodgers led the NBA in assists with an average of 10.4 per game, and played in his first NBA All-Star game. On March 14 of that same season, Rodgers tied Bob Cousy's record of 28 assists in a single game — a record that wasn't broken until nearly 15 years later.[3]
Rodgers was the point guard on the 1964 Warriors team that made the NBA finals but eventually lost the series to the Boston Celtics four games to one. In 1966 Rodgers was traded to the expansion team, the Chicago Bulls. Rodgers played the 1966–67 season in Chicago and was named NBA All-Star for the fourth and final time in his career. That same season, Rodgers handed out a then-NBA record 908 assists, which is still the Chicago Bulls single-season record.
Hal Greer was his best teammate and no mention of Jerry West (All NBA 1st team x4 2nd team x1 and Def. 1st team x 4, 2nd team x1 playing with Wilt) or Elgin Baylor (All NBA 1st team x2)?
Just an accusation without acknowledging the Hakeem vs Wilt playoff scoring advantage (even with Wilt's pace advantage)?
If there were all star quotas then please provide a link, I appreciate learning.
But even if that's true, so far I have to agree to disagree based on the evidence I've found (and still finding) Wilt had great team support relative to Hakeem. Those guys were quality players and did contribute imo.
http://www.allstarnba.es/players/most-p ... e-team.htm Or just look at Boston's guys never had more than three. Cap of three meant, as noted elsewhere 4 teams in the West meant each team got 3, 5 in the East meant an average of 2.4 per team with no team having more than 3 (so very, very unlikely to only get 1).
My post was in reference to the Warriors years but I'd assume Ardee is referring to '68 as the end of his prime his post (number 132 viewtopic.php?p=40415070#p40415070) refers to Wilt's "early days" (rookie through '63) and prime ('64 through '68). As such West wouldn't be eligible. And Baylor fwiw wasn't anywhere near his peak when playing for Wilt (much less explosive, much less efficient, much worse defender, quite a bit worse on the boards).
Thurmond can be "downplayed" (not really what was happening here) in that he was a poor fit next to Wilt. He couldn't space the floor, his rebounding was less advantageous with an already strong rebounder and his best asset was post defense and playing with Wilt he wasn't gonna do that.
Wikipedia probably isn't the best place to get a critical insight into players. This has been fairly well trodden now but ...
Watch the limited footage of Rodgers (particularly the short youtube clip) and tell be he wasn't a conspicuously left handed (and awkward) finisher at the rim. He was a good ball-handler and passer, he was, so far as I can tell, not a good defender and he was an awful shooter for a guard. The metrics really don't like him.
Arizin was a dominant player in the 50s. He was still good in the 60s. But aging plus the standard of the league improving meant he wasn't an exceptional 2nd option (though certainly still useful as a scorer/shooter).
Gola as I said great college and HS player. But he wasn't all-NBA (even 2nd team) in the 50s. And then the league got better.
As to why Ardee (and myself) didn't engage with the playoff scoring point, it's because the point of our posts were to point out that part of the post is misleading. People don't feel obliged to engage with every part of every post. But as noted in my posts Hakeem is a superb playoff player. Here is the part we felt needed attention
Also I needed to include this:
Wilt played with at least two all stars almost his whole career and had HOF coaching at times.
Supporting Casts:
Wilt 59-60 to 61-62 - P. Arizin HOF, T. Gola HOF 62-63 to 64-65 - T. Meschery, G. Rodger HOF, N. Thurmond HOF 64-65 to 67-68 - H. Greer HOF, C. Walker HOF (B. Cunningham HOF), L. Jackson, L. Costello 68-69 to 73-73 - J. West HOF, E. Baylor HOF, G. Goodrich HOF HOF Coaches: A. Hannum x4, B. Sharman x2 Total: 13 years with 1+ All Star (10 HOF), 10 w/ 2 All Stars, 1 w/ 3 All Stars, 6 years HOF coach
Followed by
Hakeem
84-85 to 86-87 - R. Sampson HOF 91-92 - O. Thorpe 94-95 to 97-98 - C. Drexler HOF 96-97 - C. Barkley HOF 01-02 - V. Carter (probable HOF) HOF Coaches: 0
Total: 8 years with 1+ All Star (4 HOF), 1 year w/ 2 All Stars, 0 HOF coach
Aside from the issues already raised the formatting is inconsistent (Thorpe is listed specifically as an all-star for one year, ditto Barkley but Meschery is for a three year stretch despite only one all-star appearance, ditto Thurmond, Goodrich is listed for a spell including years he wasn't even on the Lakers for, Baylor includes years missed through injury). Pippen (HoF) is omitted as a notable player alongside Hakeem, though Cunningham (never an all-star whilst playing with Chamberlain) is mentioned.
Not an inconsistency but somewhat misleading is a listing that includes Thurmond (in '65) and Costello and Luke Jackson as though Wilt played with a team combined of the two teams' best players in '65.
Still the big issue, as before is a list of all-star teammates as though that meant the same thing now and then.
GC Pantalones wrote:
Owly wrote:
GC Pantalones wrote:
Spoiler:
Well how come when Wilt gets to great teams his impact looks minuscule. I can't get out of my head that he raised crappy teams to average levels but didn't really help great teams as much as others here and unlike Hakeem he didn't raise bad teams to championship levels because he didn't perform in the post season.
Not sure which teams you're referring to. But my thoughts are on record here. I guess you're talking the Lakers as the elite team. That has been explained somewhat in posts but '69 is clearly on the debit side of the ledger for Wilt. Over the period he stayed there he clearly improved them significantly. I don't know to which team you're reffering to in terms of making them average. Again my vote and other Wilt voters have offered a lot of thoughts on his impact. And Hakeem is IMO, a poor analogy because he was surrounded by excellent floor spacers and defenders (and because he didn't win any more titles, albeit he did so in a larger league, and because he didn't go through the era's dominant team, or indeed as noted here his team's worst matchup - Seattle). Wilt's playoff numbers and success has been largely explained in terms of a large number of games played against Russell. But if playoffs is heavily weighted in your criteria I could understand how Hakeem looks good and Wilt doesn't so much.
When I say he made teams average I'm speaking on the late San Fransisco years. 61 on. The average level of those teams weren't too good.
I guess if you don't value postseason that much I can see this high ranking for Wilt but personally I care about how many rings I can get with a guy and Wilt has a lacking shelf life in that regard.
To clarify I don't not care how many rings I can get with "my guy", as might be read as implicit in your post. I just think playoffs (stats in particular) are subject to a lot of a factors other than player ability/typical performance level (small sample size, matchups, differing quality of opponents, gameplanning to stop you, quality of teammates/coaching to mitigate the above, injuries etc) that I don't think they always accurately represent what a player can or would typically do (particularly in terms of numbers).
Doctor MJ wrote:I'm not talking about switching time zones I'm literally talking about Wilt changing his behavior from year to year, month to month, game to game.
Game to game, eh?
Well, I leave the psychoanalysis to you, Doctor MJ. I'm not getting into that one.
Wilt entered the league capable of being the #2 MVP simply becauseof his defense, but he had a clear tendency to coast on that side of the ball which is why his teams saw such drastic ups and downs even more than the offense.
Even though he wasn't exactly Russell, his defensive value in that era was never in question.
I haven't specifically analyzed the game tape of Chamberlain's defense as of yet, so I'll refrain from comment about that side of the ball.
microfib4thewin wrote: I already saw that in the Hakeem vs Duncan thread. All you are trying to do is "Parker and Manu are future HOFers, so let's count them as playing at the HOF level for their entire career. Parker averaged 14.7 points on .468 TS and Manu averaged 9.4 points in the 2003 Playoffs. That is not HOF caliber talent. If you are still going to make sweeping generalization about the teammates of Hakeem's rivals then there is nothing more that I can say to you.
No mention of Pops, Bowen or Robinson in his earlier years and peak? What about the competition he faced at his peak vs Olajuwon?
It's not documented that Parker and Manu became stars in his later years?
We can agree to disagree.
It didn't happen to Hakeem because he didn't have as many deep playoff runs as Kareem did. Hakeem only made three Finals runs, one CF run, and three runs to the second round. Kareem made it to the Finals 9 times, CF 3 times and the second round four times, and this is counting pre-merger Finals as CF and pre-merger CF as second round.
If we grade individual player on team based feats, shouldn't we also acknowledge Kareem's team support, competition and rule advantages over Hakeem that helped enable those runs?
This actually gave me an opportunity to some research that I enjoyed though. The more I look at Kareem (who hold in high regard) the more I'm convinced Hakeem would have outplayed him if they met head to head in his prime as well.
Hakeem was mentored by Moses Malone who outplayed and beat Kareem in the playoffs in 1981, when Kareem was 33. This never happened to Hakeem at 33. (full game links in youtube)
It's probably a foreign concept to you, but there is such a thing as a player who got into the HOF because they are fortunate to play on a great team. Parish is one of them. If he had stayed with Golden State then no one would think of him as HOF worthy. The same applies to other guys like Pau and Parker where their performance would be largely ignored if they are on bad teams.
Does the same apply to valuing an individual players who are fortunate enough to have elite team support helping him win championships or make playoff runs?
While we could give credit to Hakeem's defense for Ewing's tremendous struggle in the 94 Finals(.39 TS) I don't see much credit that should be given to Hakeem's offense(.565 TS in the regular season to .556 TS in the Finals). At most, we could say Hakeem didn't get shut down by the Knicks defense, but Hakeem wasn't going wild against them either.
Hakeem put up 27 pts, 9.1 rebs, 3.6 ast, 1.6 stl, 4 blks on .556 TS% against a prime HOF Ewing and dominant defensive team which was historically great the year before with the same cast. He also destroyed Ewing defensively. This was much tougher than Duncan's or Shaq's defensive finals competition.
Shaq didn't get shut down and had difficulty going wild against a historically dominant Pistons team in 2004, putting up 26.6 pts, 10.8 rebs, 1.6 ast, 0.4 stl, 0.6 blk on .615 TS%. Ben Wallace was as good or improved from his season averages and TS% against Shaq.
Tim Duncan's team also won, but he struggled offensively against the 2005 Pistons team, shooting a much worse .471 TS% in comparison to Hakeem. Ben Wallace had slightly less rebounding, but better scoring and much better TS% in comparison to his RS averages against Duncan.
The big outlier here is 1995. He had the lowest rebounding rate in his career during that time, he hasn't hit over 100 Drtg since 1986, and he has identical DWS to 1992 despite that being Hakeem's worst season thus far. In the playoffs, he had a 108 Drtg. Shaq had a higher playoff defensive rating four times: 1995(109), 1996(111), 1998(110), 2007(109)[4 games]. Duncan had a higher defensive rating only once: 2009(110)[5 games].
The reason I mentioned all this is, if there is no doubt that Hakeem was nowhere near his defensive peak in 1995 then how do we view his 1995 run and how does that affect our view on his career as a 2 way big?
I believe Hakeem was asked to play further away from the basket, but I have to check to be sure.
I don't put much stock in accolades but during the 3 years of 93-95 when Rudy T had restructured the team offensive and defensive systems to complement Hakeem. He was clearly the best defensive player in the league during that three year span.
Again, I dislike accolades in player comparisons because other greats weren't competing in that same era with the same rules, team support or competition (Ex: Kareem, Wilt, Shaq and Duncan weren't winning MVPs with Jordan playing in the league etc.)
But to give some broad strokes of his defensive play he won DPOY twice and came in third in the voting the last year during these years.
The team defensive scheme now complemented him and he was still very impactful imo. The Rockets were still a top 5 defense (2nd in Drtg) until age 31. People usually regard his single peak in 94. I view his athletic peak in the 80s and when his roster/coaching system improved it helped him showcase it better during his 93-95 years.
Russell declined with age, Wilt declined with age, Shaq and even Duncan has lost a step. Be it defensively or offensively, it happens with every player and ATG.
Duncan has better longevity, but was worse in his peak overall. Shaq has a comparable peak but worse defensive longevity. Kareem has better offensive longevity but worse defensively. Wilt got worse offensively at the expense of his defensive longevity. Hakeem aged better defensively but got worse as he aged offensively as well.
Drtg is useful for in-team or in- league comparisons during that season. It has flaws in cross season comparisons or other teams imo.
I use it to point out who played significant minutes and lead the team or league in Drtg as it can get muddy in other context imo.
I'll post the Drtg flaws example later as I'm short on time. But here's Defensive Win Shares and PER's flaws in comparisons for now:
Spoiler:
I'll be posting to others responses as time and notes permit. But I had this on hand so I'll post it regrarding comparing players defensively based on Winshares, which is a team based stat that is inflated by team support or lesser competition (like Mikan) or PER which is an offensive stat.
This I disagree with for comparison purposes as PER is a flawed stat used primarily for grading offensive players and is especially poor when comparing great defensive players.
Problems With PER
PER largely measures offensive performance. Hollinger freely admits that two of the defensive statistics it incorporates -- blocks and steals (which was not tracked as an official stat till 1973) -- can produce a distorted picture of a player's value and that PER is not a reliable measure of a player's defensive acumen.
Win-shares are also bad because they are team based stats that can be dependent of roster strength to increase total team wins and boost win-share totals. They also don't account for beating lesser competition.
WS/48 Win Shares Per 48 Minutes (available since the 1951-52 season in the NBA); an estimate of the number of wins contributed by the player per 48 minutes (league average is approximately 0.100).
Please see the article Calculating Win Shares for more information.
Calculating Win Shares 1.In James's system, one win is equivalent to three Win Shares. In my system, one win is equivalent to one Win Share.
2.James made team Win Shares directly proportional to team wins. In his system, a baseball team that wins 80 games will have exactly 240 Win Shares, a baseball team that wins 90 games will have exactly 270 Win Shares, etc. In my system, a basketball team that wins 50 games will have about 50 Win Shares, give or take.
Criticism of win shares One criticism of this metric is that players who play for teams that win more games than expected, based on the Pythagorean expectation, receive more win shares than players whose team wins fewer games than expected. Since a team exceeding or falling short of its Pythagorean expectation is generally acknowledged as chance, some believe that credit should not be assigned purely based on team wins. However, team wins are the bedrock of the system, whose purpose is to assign credit for what happened.
Win shares are intended to represent player value (what they were responsible for) rather than player ability (what the player's true skill level is).
In other words they can be inaccurate with the playoffs especially bad. Look at the yearly playoff win-share leaders below, they clearly don't indicate the best player accurately every year. ex. Jamaal Magloire in 02', Marcus Camby 99' , Hardaway and Smith over Jordan in 97'-98' etc.
Another example is (in 2013) Tiago Splitter is 8th in the league for WS48 over Blake Griffin, Russell Westbrook and Duncan himself. Chauncey Billups was over LeBron in 07-08', I could go on. Check the top 10 yearly RS WS48 link below.
Advanced stats are often based on weighted box score stats too (which weighting is of the creators subjective discretion), so box scores while flawed can actually be more accurate at times and vice versa. Basketball has so many more variables per play than Baseball where win shares originated from.
They are useful when your understand their context and flaws for comparing players within a team within a season, but not different team players in different seasons imo. They are bad for comparing defensive impact, let alone across eras imo.
I will accept the articles for now. I will need to do some reading to get more information on Hakeem's historical context.
Alright, I'm glad you are willing to acknowledge good opposing points. Good substance filled exchange on your part as well.
Doctor MJ wrote: Colts please stop referring to this without some acknowledgement of the discrepancy between studies.
You're using non prior informed RAPM and both pure APM and prior informed RAPM give Garnett the edge over Duncan and make Garnett vs Shaq look quite close.
How can you attack my post without acknowledging that your spreadsheet uses a NPI RAPM that has only 30% of the season completed? Shouldn't it make you curious why Duncan is ahead of KG in NPI RAPM every year from 01-04 yet KG has a higher prior informed RAPM that season? J.E. has acknowledged that the early seasons of RAPM (02-06) are incomplete yet you take it as the gospel. It should make you reconsider your stance when I point out discrepancies like that. It would make no sense that KG is ahead of Duncan in 2004 if Duncan has him beat every year prior. That only happens with incomplete data.
All I did was ask you to acknowledge the discrepancy, not any of the more violent things you just accused me of.
Okay so for the room:
It's really unusual for the non prior informed and prior informed RAPM to disagree like this. In any given year sure, but with any kind of consistent behavior on the part of the players over many seasons they basically always tell a similar story. And even here we're not talking anythjngtoo crazy, everything agrees these two are superstars.
That the side behind opposing parts of this particular comparison though is some thing we can't just ignore.
So basically as I understand it here are the two proposed explanations:
1) RAPM looks to smooth out outlier data. The smaller the sample size, the more this has an impact which is precisely why it is preferred over pure APM in terms of reliability but it's still better if you can glean expectation from a greater sample size, and the preferred way to do this is by using a prior.
So hypothetically were we to see a player who did worse by RAPM with no prior than he did by pure APM, but when you included the prior he looked better again, one explanation is simply that the smoothing out process was wrongly calling the player lucky.
This particularly makes sense in Garnett's case because his teams performance when he went our was so godawful that it's understandable to chalk it up to luck to some degree.
2) a second explanation is that the data itself used for the studies wasn't Ty the same. colts is saying this and also he's saying that the data he stands by is complete where the data of the prop informed study is not.
Here's the thing though: the same trend is happening repeatedly. I forget how many years where the two methods disagree in this comparison but it's WAY high. In light of this to me it makes one explanation stronger and the other weaker.
for the incomplete data to be the true culprit here, that would mean that in every distinct season the data that just happened to be used was the part of the season that Garnett looked better and the part that was list just happened to be whereDuncan looked better.
Something like that happening randomly once is plausible, but repeatedly? No way that's the most likely explanation IMHO.
