RealGM Top 100 List #6

[Doctor MJ]

So quick thoughts:

When I see people talking about "advanced stats can be wrong" it says to me the person really just isn't comfortable with data analysis, bra cause no one who is used a phrase like "advanced stats" as part of their advocacy.

It's no sin to be uncomfortable, and anyone who says you can't have an opinion unless you are a monster data miner is a fool. It's also quite reasonable to say "look you're just not reaching" me with this line of argument. However when you find yourself rebutting extremely specific statements with vague "can be wrong" statements I think you really ought to try to get better acquainted with these newfangled things.

Second: it is frustrated to see people say "all you're saying is advanced stats" when I saw drza quite recently go into vast depth breaking down Garnett through the years in a granular detail that hasn't been done for most of the other candidates by anyone else. It really seems to me that the pro-Garnett side here has flooded the conversation with arguments here.

I certainly don't expect that to mean everyone is convinced, but there's a difference between being unconvinced and disparaging the efforts. The latter to me, in the face of what all I've seen, seems to indicate that just the opposite is true: that the reality is for a lot of people here right now there's just not anything that can really sway them any more than they already are away from their prior conception.

And you know that's not anything too awful. It's fine. Just try to keep your mind open in the future and ask questions if there are things you don't understand.


Sent from my iPhone using RealGM Forums

[Baller2014]

When I see people talking about "advanced stats can be wrong" it says to me the person really just isn't comfortable with data analysis

Exactly the response I expected. Basically if we don't agree with you, then we're wrong. You can paint it in different words, but that's basically what you just said. Guess what, I don't agree, and I'm in very good company.

[therealbig3]

IDK what else you need, drza's post wasn't solely advanced stats...he used them to advise his qualitative analysis of Garnett. ElGee did the same thing.

Posting stuff like "Garnett didn't carry so-and-so team to more wins" or "Garnett wasn't a go-to scorer" in light of what they posted just seems like very rudimentary and shallow stuff. It's extremely unconvincing. How about actually responding to specific points that they've made and bringing up appropriate counter points that they can respond to without feeling like their time is being wasted? Because they clearly spent a lot of time researching and compiling those posts, only for some people to make cursory statements that basically reveal that they either didn't understand the posts, or didn't bother reading it.

I mean, if you need another example, there's another one in this thread: ElGee is a Celtics fan that likes Bird a lot, and he's advocating for him over Magic (which tends to be a minority opinion)...but he still gave appropriate context in his analysis of 1980 Bird and clearly pointed out that saying things like "Bird took a 29-win team to 60+ wins by himself in his rookie year, clearly he's super awesome!" is misleading and inaccurate. And then someone proceeds to post pretty much that exact statement almost verbatim in a subsequent post.

It's just frustrating to see people not even try to debate certain points and continuing to stick with their preconceived notions. Isn't discussion the whole point? If someone has a different opinion than you, don't just do the message board equivalent of shouting your opinion over them...respond to it and articulate specifically why you disagree.

[Baller2014]

Drza's post tells us KG was a great player, even excluding the advanced stats. But he's not being compared to merely good players here, he's being compared to great ones. Take the section Drza wrote on 2002. He didn't address any of the support cast issues I spoke about, and why the team only won 50, not really. All he focused on was how KG was a great defender that year. The thing is, I don't disagree he was a great defender. I want to know why his supposedly Duncanesque impact translated to only 50 wins and a first round exit.

I noted I disagree with Elgee's downplaying of Bird's role in 1980, if anything the support cast might have been worse in 1980 than 79. It was probably roughly equal.

[therealbig3]

I mean, it's literally right there in his post...the Timberwolves started off super hot out of the gate, going 30-10, using a novel defensive strategy that the rest of the NBA wasn't used to seeing. Yes, eventual adjustment by the rest of the league is a valid assertion. And, their starting PG gets hurt, which means Billups, who wasn't as good of a PG, takes over, while KG has to take on more offensive responsibility, which causes his defense to suffer, which hurts the team more, because he was the one that held the defense together. They only go 20-22 the rest of the way and end up with 50 wins as a result.

And once again, he explains in extreme detail how the entire team's defense falls apart, how the zone concept that Saunders was using had run its course, and how the Mavs offense was perfectly designed to torch them...meanwhile, Dirk's monstrous series actually had very little to do with KG's man defense, because KG was barely guarding him.

I understand he's being compared to other great players here. I wish I could get some posts like that about the other great players that really break down their careers in extreme detail, as well as ElGee's last few posts about Garnett that do a great job of breaking down KG's skillset and why exactly it's so special. I mean, for example, a lot of your Duncan posts were merely a bunch of player assassinations that unfairly trashed his supporting cast just so that Duncan looks as good as possible, all the while completely ignoring the impact of Popovich, or the stability of the Spurs as a franchise in general, and just how good of a situation he found himself in.

[90sAllDecade]

I didn't include 1986 because I wanted it to be used for nefarious purposes and discredit Olajuwon. No, I originally used this for Ewing and picked 11 seasons for him, and I went with 11 for Olajuwon for a fair comparison. I didn't want to use seasons when the guys were too old or too young, but it's trickier with guys who have interesting careers like Shaq or Olajuwon. Maybe I'll include careers with different years.

I already added in the season and analyzed it quickly.

1986 was one of his best seasons by this metric and it's on par with Shaq's 11 year average. However, his usage doesn't climb as he faces better defenses. Adding in 1986, his coefficient for ORtg is now 0.98. So it's still almost exactly 0.1, which is the average you'd expect.

No weighing. Right now it's only a simple linear regression. When I get more data I'll dump it into R and mess around with it more. I could weigh the playoffs more. I said above it's playoffs included, so it's every regular season game and every playoff game.

What teams were used? Maybe this wasn't clear but it's everything. I'm taking his complete gamelogs and linking his performances to the opponent's defensive rating. It's every team.

Some residual analysis will be useful soon, but it's still a lot better to treat this as a continuous variable rather than the simple summaries people use.

Hmm, curious why you didn't include defensive rating and used 2005 as the cutoff as going past that hurts Shaq. Why not do it for their whole careers for their first 17 years?

Anyhow, I'm working on this data myself. It will be on an averages scale, since I've ventured into math like this a few times but I'm no expert.

[Baller2014]

I mean, it's literally right there in his post...the Timberwolves started off super hot out of the gate, going 30-10, using a novel defensive strategy that the rest of the NBA wasn't used to seeing. Yes, eventual adjustment by the rest of the league is a valid assertion. And, their starting PG gets hurt, which means Billups, who wasn't as good of a PG, takes over, while KG has to take on more offensive responsibility, which causes his defense to suffer, which hurts the team more, because he was the one that held the defense together. They only go 20-22 the rest of the way and end up with 50 wins as a result.

I told you, I read this. But it's not an explanation, or at least not a satisfactory one. His point guard getting hurt clearly wasn't the issue, as I pointed out the team was 20-12 with Brandon and 30-20 with Billups. So they were marginally worse, a 51 win pace v.s a 49 win pace, but the sample size is small enough it could easily be a push one way or the other. Saying "the rest of the NBA adjusted" is meaningless. Every team faces that every year, and we call the final result how good you really were. Houston was undefeated by definition on their 22 game win streak in 2008... it didn't mean that cherry picked hot streak was in any way a fair reflection of their talent as a team, to find that we look at their overall performance that year. I mean, that's why they're called averages. The same is true here.

The reality remains that KG had a good support cast that year (and other years, but 2002 is the most obvious), when we are comparing it to what the guys he is being compared to had (Duncan's rubbish in 01-03, Bird and the 29 win Celtics, Kareem and the 27 win Bucks, Lebron's weak Cavs, Dr J's 76 Nets, Walton's Blazers, etc, etc). Despite that, KG's teams invariably had worse outcomes, and that counts against him. A lot. You can make a bunch of excuses for every year he failed, but at the end of the day it rings hollow. He had plenty of chances to carry a team like the guys he is being compared to, and despite more than reasonable opportunities (like the clearly decent 02 support cast) he never could do it. You can believe that every year he had a bunch of excuses, or you can take the (much more reasonable) position that he just couldn't do it. Your call.

[90sAllDecade]

On a side note, how do people compare Garnett with Dirk?

[therealbig3]

We can also double-check and see if drza's explanation is backed up by the stats. He asserts that the defense was adjusted to, as well as KG's increased offensive responsibility, was responsible for the team's 20-22 finish. Also, drza suggests that the team offense started to struggle after Brandon was replaced by Billups as the starting PG.

Minnesota defense in the first 40 games (30-10): 104.7 DRating vs 104.9 ORating = -0.2 defense
Minnesota defense in the last 42 games (20-22): 105.9 DRating vs 104.1 ORating = +1.8 defense

So yes, the defense did fall off, by 2 points.

Minnesota offense after Brandon's injury (18-16): 109.4 ORating vs 104.8 DRating = +4.6 offense
Minnesota offense before Brandon's injury PG (32-16): 108.7 ORating vs 104.3 DRating = +4.4 offense

So in fact, the offense plays slightly better with Billups as the starting PG than with Brandon...but it coincides (it was the last 34 games of the season) around the time when Minnesota's defense fell off...overall, the injury to Brandon certainly hurt them, because they went from a +4.6 team with him to a +2.8 team without him (not exact, because it's 42 vs 34 games, so there are 8 games unaccounted for, but it's an estimate). This would perhaps be because of drza's assertion that Garnett had to spend more energy on offense, resulting in a slippage on defense.

[Purch]

When I see people talking about "advanced stats can be wrong" it says to me the person really just isn't comfortable with data analysis

Exactly the response I expected. Basically if we don't agree with you, then we're wrong. You can paint it in different words, but that's basically what you just said. Guess what, I don't agree, and I'm in very good company.

This is a big reason I wanted Shaqattack in on this project. I've known him from this forum and other forums and there's no ones opinion who I respect more on the game of basketball. Because his methadology doesn't rely on advanced stats, it relies on a genuine understanding of the game, and the history of the game, which can actually be overwealming. Thats no insult to other posters, but I feel they're way to willing to let every new advanced stat that comes out sway their perception dramatically, and then struggle to understand players from history where data isn't available. His old youtube channel reflects just how many games he owns, and just how much game footage he's watched over the years. His understanding of the game and methods of dissecting it was a huge reason why I built up my big collection of nba games, and I continue to break down video footage.

I'll always recpect posters who feel that advanced stats should be the foundation of every argument, but I put higher value on arguments that clearly demostrate hours dedicated to the history of the game and the current game.

It's not that advanced stats are wrong, it's that they don't paint a convincing argument to me.

[Baller2014]

We can also double-check and see if drza's explanation is backed up by the stats. He asserts that the defense was adjusted to, as well as KG's increased offensive responsibility, was responsible for the team's 20-22 finish. Also, drza suggests that the team offense started to struggle after Brandon was replaced by Billups as the starting PG.

Minnesota defense in the first 40 games (30-10): 104.7 DRating vs 104.9 ORating = -0.2 defense
Minnesota defense in the last 42 games (20-22): 105.9 DRating vs 104.1 ORating = +1.8 defense

So yes, the defense did fall off, by 2 points.

Minnesota offense after Brandon's injury (18-16): 109.4 ORating vs 104.8 DRating = +4.6 offense
Minnesota offense before Brandon's injury PG (32-16): 108.7 ORating vs 104.3 DRating = +4.4 offense

So in fact, the offense plays slightly better with Billups as the starting PG than with Brandon...but it coincides (it was the last 34 games of the season) around the time when Minnesota's defense fell off...overall, the injury to Brandon certainly hurt them, because they went from a +4.6 team with him to a +2.8 team without him (not exact, because it's 42 vs 34 games, so there are 8 games unaccounted for, but it's an estimate). This would perhaps be because of drza's assertion that Garnett had to spend more energy on offense, resulting in a slippage on defense.

Like I said, you can play that game with most teams- pick a sample in their season when they got off to a good start (or the reverse) then blame something for the fall off (or improvement). If we take the Pacers only off the start of the season they're a top 3 team this year, but as the season went teams adjusted and they fell off. That doesn't mean we give the Pacers brownie points and say they were "really" a top 3 team, because the NBA season is 82 games long. Other teams all were "adjusted" to, and they held up. Are we supposed to punish them because they didn't fall off? Absurd.

But wait, it turns out that when the Pacers played better some advanced stats said they were a better team! I am stunned. Obviously, every team on a hot streak ever will be posting better stats when they are on a hot streak than when they are not. It's also just illogical to cite it here too, because if the Wolves have Brandon starting and Billups coming off the bench, logic tells us that of course they'll play better... they've got 2 good players instead of one. That really gets us nowhere in the discussion of whether KG had "enough" on his team to expect a better results, and on that discussion all I see are excuses as usual. Even with zero games of Brandon he should have had enough help to do more if he's really of comparable impact to Duncan. Needless to say he is not.

[rich316]

I had to miss a few votes to go on a great camping trip with my fiancee, but I'm glad to be back!

My vote for #6: Shaquille O'Neal.

Watching Shaq sh*t on various playoff opponents was my first memorable experience as a basketball fan. This definitely colors my opinion a bit, as I never had a chance to watch Bird or Magic play. Still, I don't think it's unreasonable to sustain Shaq's ranking above them based on a real analysis. In the cases of Bird/Magic, you are basically looking at a 5 year window when either of them have the ability to lead you to a title and win multiple MVPs. Shaq did that for 10 years. His peak started basically as soon as he was drafted, and lasted until 2005. His 2000-2002 stretch is a clear notch above anything Bird/Magic achieved, IMO. While his defense wasn't all that great after his athleticism started to go, at his peak it was significantly better than anything the 80's stars did. For the defensive edge to be cancelled out, you would have to see a substantial edge on offense for either of them over Shaq, but you don't. I don't think I've ever seen a player score more easily than Shaq in his prime. He dunked on teams like he was walking his dog. He was also the best big man passer I have ever seen. His bear-claw sized hands let him whip the ball around the court like he was tossing a tennis ball. He did all this while his opponents were basically trying to jump on top of him, tackle him, put him in leg locks - anybody remember Mark Madsen?

Yes, the biggest knock on Shaq is his inconsistency/durability. When you're comparing him to Bird, however, that doesn't seem like such a drawback. Imagine if Shaq had injured himself in a bar fight during a playoff run in his prime - the cries of "selfish" and "cancer" would be defeaning. As ElGee has nicely argued, Magic's prime didn't really get started until 1984.

If you are arguing for Hakeem, it seems to me the main virtue of Olajuwon is his sky-high peak. Unfortunately for him in the comparison, Shaq's peak matches right up.

[trex_8063]

I mean, for example, a lot of your Duncan posts were merely a bunch of player assassinations that unfairly trashed his supporting cast just so that Duncan looks as good as possible, all the while completely ignoring the impact of Popovich, or the stability of the Spurs as a franchise in general, and just how good of a situation he found himself in.

Statements like the bolded ignore Duncan's contribution to the "good situation" he's found himself in, a contribution I've attempted to point out on multiple occasions. :

.....Now I know a lot of people will sing the praises of Gregg Popovich for some of these successes (as well they should). But I want to give Duncan his portion of the credit for the attitude and culture that exists in San Antonio. When new players (veterans, rookies/young players, emerging all-stars, etc) come to the Spurs team and they see the team’s superstar---likely the greatest player they’ll ever play with---conduct himself with poise, humility, a lack of drama or theatrics, and always a team-first mentality…….that goes a LONG way to creating the kind of team culture that Pop often gets the credit for.....

.

[TrueLAfan]

Why outside of his control? Somebody forced him to all these intercourses?

Was someone else forced to retire that I’m not aware of? The players that were vocally against having Magic come back were against having basketball players people tested for any infectious diseases

Magic's sex life was the reason why he retired, not opinions of other players or anything else, ergo you can't say he retired because of factors outside of his control, because these factors were totally under his control - the only problem was that he wasn't able to control himself. Sad but true.

Magic’s “sex life” has nothing to do with this. Magic Johnson was forced to stay out of the league in 1993 because he was HIV positive. People of all types, with all types of “sex lives” contract the HIV virus. The league publicly stated they were fine with Magic’s return in 1993 (see David Stern’s interview and commentary on it). Players let their fears—some of which were medical, some of which were, frankly, related to not liking and/or having Magic lead the Lakers in the WC—override league policy and medical fact and consensus. If you don’t know this, look it up.

[Basketballefan]

I just want to say i don't get the lack of love for magic. Russell is praised for being the best defensive player ever, so why don't we do the same for magic for being the greatest offensive player ever?? I'm appauled that Magic will likely end up 7th or lower.

[shutupandjam]

This is fascinating:
http://shutupandjam.net/nba-ncaa-stats/ ... ed-impact/

One critique: when I made my own all-time ranking based on stats alone, I found the David Robinson/Karl Malone problem. They were ranked really high, higher than in reality. But it's because we often just use regular season stats even for things like estimating championship odds. Any way of using playoff stats, or an approximation thereof, for this odds method? Like estimate a player's average drop in impact in the playoffs and apply it to every season. By the way, Kareem led the league in that estimated impact ten straight seasons. I know the 70's were an odd decade, but jeez....

In the old version of estimated impact I had playoffs, but the issue is team strength. Raw net rating for every team is easily accessible, but it doesn't truly reflect each team's strength because you're not playing the 'average' schedule in the playoffs. I'm actually gonna go year by year and calculate adjusted net rating based on opponent regular season strength. Then I'll look at average PS drop for each player and apply that to "Expected Titles." That should definitely serve to temper guys like Robinson and Malone and boost guys like Hakeem and Russell.

[NO-KG-AI]

It's just frustrating to see people not even try to debate certain points and continuing to stick with their preconceived notions. Isn't discussion the whole point?

This has been my biggest problem, and reason for posting less and less over the years. I've been on here a long time and learned a LOT because I came in with an open mind. Most people aren't, and way too many of them are allowed to vote in these things, when it's clear they have no willingness to change their opinion, no matter how much evidence is thrown in their face. It's tiresome at this point, that people spout simple, unsubstantiated, and even things that are proven completely wrong like it's fact, and discredit the amazing analysis some people put forth on here.

[ElGee]

realbig3 -- I overlooked something that I thought you'd like in my Bird post. After the bar fight in 85, while the Celtics offensive numbers fell way off, the SRS dropped 1.7 points only because the defense was better. I typically always check 4-factor variables instead of just looking at these numbers in a vacuum, but in this case I skimmed right over something that suggests a systemic/strategic change simply related to playing Philly/LA that impacts the oRtg/dRtg split:

Pre fight team rebounding: 35.9% OREB% | 66.3% DREB%
Post fight rebounding: 28.6% OREB% | 71.5% DREB%

While Bird's play/shooting does fall off a bit IMO, I do think it's to a small degree, and probably not fair to conclude Boston's offense "fell apart" as I have previously suggested.

[Mutnt]

For my #6 vote, I'm going with: Shaquille O'Neal

Basically, I've had Shaq before Duncan and even before Wilt for that matter, so it's logical I'll keep voting for him until he gets in.

Everything was explain in the last thread but to sum-up: Shaq is basically a Wilt-type player. Generally less potent on defense because he rarely extended outside the paint and was often times lazy (as was Wilt tho, but ok). Compared to Wilt he is a more dominant and efficient scorer. Also an all-time great rebounder and a solid passer. He's got a top 5 peak and a great prime. Some decent longevity years in the mix as well, doesn't hurt.

[colts18]

There's absolutely no way these numbers are close to right.

First, KG had to average about 25-15-5 through the 12 games of the first two series. What are the next 2 series that you speak of? He'd have to average like 7 points and 3 boards over the another 12 games to get averages that low and he never did that

Are you counting their series as teammates in Boston as the other 2 series? Wait...Shaq only played in 2 of their 9 games that year...how did you handle that in your averages?

And there's no way that Shaq could be 3 - 1 and KG be 0 - 4.

I think you need to do some revisions here

I'm referring to the 4 playoff series in the years where they faced the same playoff opponent.