RealGM Top 100 List #8

[DQuinn1575]

Complaining about Rashard outscoring Dwight in RAPM is like saying "DeAndre Liggins averaged 72 points and 36 rebounds per 36 minutes! The stat says he's better than LeBron! Per-minute stats are useless". Every stat gets wonky with small sample sizes.

Liggins played 1 minute. Rashard in Orlando in 2009 played 2,859 minutes.
I'm not close to reading every word here - but that doesn't look comparable for sample size arguments.

[Baller2014]

If anyone is curious I believe the vote at the moment is Bird 3, Magic 3, Hakeem 2 and KG 1. In my mind that's the order for the next 4 candidates too, and then we start to get into slightly dicier territory with Dr J v.s K.Malone, etc.

[DQuinn1575]

Again, at the risk of being a broken record, at some point we are going to have to deal with George Mikan as wll. If your key is "overall impact over a sustained period of time," then Mikan may be the strongest candidate left assuming the time we are looking at him Is enough to be considered "sustained." At the same time there are huge questions needing to be answered about him but soon (after Hakeem? After Moses?) we need to a least look at his candidacy.

Forget about any era adjustment for 2 minutes:
MIkan played 5 NBA seasons ages 25-29 with an aborted 1/2 season comeback at age 31 (We are not counting BAA).
He was best player on 4 championship teams - 1st team All-NBA all 5 years. There was no MVP award but it's pretty fair to say he would win about 3.

So we have Top 7 credentials of 3 MVP/4 FInals - but he clearly would fall behind that level with only 5 seasons.

We go Magic/Bird/Hakeem/Kobe in some order 8-11 (that was their order last time) - with longevity he really doesn't have a case against any of them.

The next group from last time is
Erving/Karl Malone/Garnett/Oscar/West/Moses/Dirk/Pettit/Barkley/Robinson/Wade

I can talk about him then at 11 or 12.

My 2 minutes above are up.

When we do talk about that group, era and competition have to be talked about regarding 50s,60s, and ABA.
But I will wait to then and not take away from the matter at hand - #8.

[An Unbiased Fan]

Spoiler:

Pretty amazing that there were 30 minutes every game that Manu could be used in lineups catered just to his skillset, and that he could make those lineups amazing. Even more amazing that those lineups weren't nearly as good when Manu wasn't on the floor. And that even though Manu went from the bench to starting and back several times, and just about every player on the Spurs over the past 10+ years has had plenty of minutes spent with him and another large sample of minutes without him, Pop still only put him in the huge variety of situations where he would succeed and none of the ones where he would fail. Glad we could find an explanation for 10 years of overwhelming data that boils down to "eh, it's probably a fluke of a lucky situation.... er, hundreds of different lucky situations"

What's the point of misrepresenting my post. I didn't I say, "eh, it's probably a fluke of a lucky situation.... er, hundreds of different lucky situations"? I said the exact opposite. I don't believe RAPM is flukey, but instead gives a fairly accurate representation of how effective certain lineups were. The problem is you can't extract individual impact from it, so using it in comparisons is pointless.

I also said "Pop has always typically put Manu in units where his skillset can be utilized most effectively", not that Pop only used him "in the huge variety of situations where he would succeed and none of the ones where he would fail". This speaks to coaching 101.

Spoiler:

Orlando in 2009 has been discussed multiple times, and I wish you'd stop bringing it up, because anyone who knows RAPM knows not to use it in that situation, at least once they're made aware of the details. It's a situation with a ton of collinearity, where the difference between the Dwight/Gortat pairing and the rest of Orlando's lineup basically boils down to a few dozen minutes per season. Complaining about Rashard outscoring Dwight in RAPM is like saying "DeAndre Liggins averaged 72 points and 36 rebounds per 36 minutes! The stat says he's better than LeBron! Per-minute stats are useless". Every stat gets wonky with small sample sizes.

The thing is, this collinearity is incredibly rare. Orlando in the Dwight/Gortat years is the single go-to example for where collinearity can occur. In most cases, we have an incredibly varied sample of lineups to look at. Just like we recognize that points per minute isn't a good measure to look at for DeAndre Liggins, but still use it to evaluate Kevin Durant, we can recognize that RAPM has a tiny sample size for comparing centers to non-centers on the 2009 Magic, and still make use of it in the 99% of the time when it does have an adequate sample.

Did you actually read my posts, because my reasons for bringing up Lewis weren't to "state" that his numbers were "wonky". You are literally arguing against points I didn't make.

I don't think RAPM results are wonky, I think the results are very good....for lineup evaluation. My issue isn't with RAPM, but how its being misinterpreted. I've said this more than a few times since its been in discussions.

When you say,"we have an incredibly varied sample of lineups to look at", my whole point is that you should recognize that RAPM represents that very thing...lineups. Not individual impact. Yet again, you would need different data to be used as a base like Synergy to even attempt that.

Spoiler:

You're indicating a total lack of knowledge of how APM/RAPM is calculated when you say this. Each player gets a score, and the sum of the scores of all players on the floor is taken into account. If I play with Tony Allen, who has a +2 individual defensive score, and the team has a +2 defense, and then I play with James Harden, who has a -2 individual defensive score, and the team has a -2 defense, my score will be the same either way. I come out to a neutral impact. This is a bit of a simplification, but it'll do for now.

If there was some huge shift in strategy as rotations changed, this might make sense. If Popovich went around saying: "The starters are benched, so now only play defense. Just screw around and do whatever you want on offense," and then he turned around and said "Ok, starters. Just let them score. All I care about is how much you score," and then after that he said "Now, Manu's out, so just play awfully on both ends so he looks like he's making a huge impact by comparison", then I'd get it. It would make sense that "But.... LINEUPS!" is your rebuttal to every single thing anyone says about RAPM. But as it is, you just have basketball lineups. Every lineup tries to maximize their level of play on both ends. Sure, you have defensive lineups that are defensive lineups because they feature better defensive players, but APM/RAPM adjusts for that. That's what the A stands for.

Oh boy. So again, I have to wonder if you even read my posts, or just went to the boilerplate RAPM response of "You're indicating a total lack of knowledge of how APM/RAPM is calculated".

1) The notion that a coach will tell his players, "The starters are benched, so now only play defense. Just screw around and do whatever you want on offense", is just dismissive hyperbole. That was never my inference. I'm baffled that coaching rotations seem so foreign to you. Basketball is a game of matchups, and every coach favor certain units for certain situational matchups throughout a game. Could be small ball, or 3pt units around a big to space the floor, or so on and on.

2) APM/RAPM can't give "individual scores". All data being calculated is based on what 5 players did on the floor for each possession. You can find lineup trends, but nothing more. Which is fine.....as long as people don't try to claim it represents individual impact, and in individual comparisons.

NOTE: I want to drop RAPM debates because they overtake threads, and frankly, I don't think RAPM backers are very willing to question it. It's like arguing religion to a devout follower, no minds will be changed.

[Purch]

I actually predict that Kobe will pick up some major momentum around #10.

Honestly of the 3 candidates left, For me it would probally go, Bird, Magic and then Hakeem

[An Unbiased Fan]

Again, at the risk of being a broken record, at some point we are going to have to deal with George Mikan as wll. If your key is "overall impact over a sustained period of time," then Mikan may be the strongest candidate left assuming the time we are looking at him Is enough to be considered "sustained." At the same time there are huge questions needing to be answered about him but soon (after Hakeem? After Moses?) we need to a least look at his candidacy.

Longevity will push him into the late teens for me. He simply didn't play enough NBA years to be this high.

[Baller2014]

I actually predict that Kobe will pick up some major momentum around #10.

Honestly of the 3 candidates left, For me it would probally go, Bird, Magic and then Hakeem

Yeh, my feeling from the pre-list threads and earlier discussion was that he'd be in by about #15. I'm looking forward to the debate between him and Karl Malone.

[Colbinii]

Yeh, my feeling from the pre-list threads and earlier discussion was that he'd be in by about #15. I'm looking forward to the debate between him and Karl Malone.

Don't think Malone/Kobe is much of a debate, so I can't wait to hear Malone's side!

[RSCD3_]

Now, you might take that point and say "Okay, and isn't that something worth talking about as something Dirk might be better at?". Absolutely it is, but it's also a part of other things. We can absolutely talk about it, but it's not really very useful to try to split and split and split things down for the sake of finding some small little thing that Player X is better at imho.

One of the places where we are disagreeing is that some of us don't see Dirk's ability to distort defenses by his mere presence on the court "some small little thing". If you feel like it is, then that explains a big part of our disconnect.

Exactly and a major selling point of shaq was his goat level distortion so we cant just throw it out the window now

[ardee]

Now, you might take that point and say "Okay, and isn't that something worth talking about as something Dirk might be better at?". Absolutely it is, but it's also a part of other things. We can absolutely talk about it, but it's not really very useful to try to split and split and split things down for the sake of finding some small little thing that Player X is better at imho.

One of the places where we are disagreeing is that some of us don't see Dirk's ability to distort defenses by his mere presence on the court "some small little thing". If you feel like it is, then that explains a big part of our disconnect.

Yeah, there are few things more valuable in basketball than a defense-distorter on the level of a Dirk or a Kobe or a Shaq.

It basically means the opposing team's defensive gameplan is completely shot to pieces.

Doctor MJ, I'm happy to dig up film of the 2011 Playoffs and break down for you exactly HOW MUCH Dirk was doing for the Mavs offense without even touching the ball.

The number one factor for the Mavs title was not Dirk scoring 28 ppg on 60+% TS, but rather the fact that BECAUSE he did that, opposing defenses gave a LOT more space to his teammates.

If this ability of his lead to a title, I don't see how it's a 'small thing'. It's arguably the single most impactful thing that either Dirk or KG does.

[shutupandjam]

A couple of questions for the KG supporters:

1. Prime vs. prime, do you think KG was the better defensive player than Robinson? Why?

2. Prime vs. prime, do you think KG was the better offensive player than Robinson? Why?

3. Suppose for argument's sake Robinson's prime was better than KG's. How much of a longevity advantage is 1996-2000 for KG?


Some numbers:

NPI RAPM at the same age, starting with the first year we have for Robinson:
Age 32: Robinson +3.8, Garnett +4.4
Age 33: Robinson +6.1, Garnett +2.1
Age 34: Robinson +3.6, Garnett +4.0
Age 35: Robinson +4.0, Garnett +4.4
Age 36: Robinson +3.0, Garnett +2.4
Age 37: Robinson +3.1, Garnett -1.0

Late career Robinson easily stands up to late career Garnett, and remember, Robinson is the one who supposedly declined more sharply past his prime (or more specifically past his 1997 injury).

Interesting note: In 1999 and 2001, Robinson had higher defensive npi rapm numbers than any year in Garnett's career.

We don't know a whole lot about DRob plus minus wise before 1997, but 1992 is an interesting case since he missed his team's final 14 games. Before those games, and with Robinson, the Spurs had a net rating of +4.5. During those games without Robinson, the Spurs were -5.1. That's a swing of 9.6.

Robinson's box score numbers are far superior to Garnett, but I think we all know this. Just to throw a few numbers around though:

Career estimated impact: Robinson +5.8, Garnett +3.6
Career WS/48: Robinson 0.250, Garnett 0.185
Career PER: Robinson 26.2, Garnett 22.9
Career USG &TS%: Robinson 26.2% & 0.583, Garnett 25.2% & 0.547

So what I think it comes down in the DRob vs. KG debate are the the second and third questions I posed (I think most will admit that Robinson was at least KG's equal on defense).

Does KG have the offensive advantage? He's a better passer and better at spreading the floor, but he's not nearly the go-to scorer Robinson is. Robinson has 5 seasons with a pace-adjusted points/36 higher than KG's best season and 7 seasons better than KG's 2nd best. Also, Robinson has 10 seasons with a higher ORB% than KG's best. That may not be an entirely fair comparison because Robinson played closer to the basket, but at worst it serves to offset the advantage KG creates with spacing.

And did KG really do that much in 1996-2000 to give him a huge longevity advantage? I'm not sure. But for me, Robinson's peak advantage is greater than KG's longevity advantage.

[Doctor MJ]

I saw an argument for Dirk over KG that to me looked like this:

1.Dirk is objectively the better offensive player and here's proof: PPG, TS%, etc.
2. Sure Garnett's better on defense, but by how much really?

Now, my first quibble here is that the 'proof' about Dirk's superior focused entirely on scoring, which is only one part of offense. I certainly don't deny that in this case that one facet of offense is enough to give Dirk the overall nod on offense, but it's the imbalance of the argument in striking because Garnett's not only better on defense than Dirk, he's also better at all the other facets of offense, yet the weight of the argument seemed to lump the 90% of the game that's not individual scoring into something that can be waved away.

That was me, and you're skipping over the substance of the argument.

When someone is giving up approximately 17-20 ORtg points on higher volume, the issue becomes bigger than scoring. A guy like Dirk, rocking 30+% USG on 120+ ORtg, is going to be a virtual black hole on the court. Rewatch the 2011 Playoffs, he sucks defenders toward him from all sides of the court. Even if he doesn't have the ball, defenders are always splitting attention because his mere PRESENCE on the court is a threat. It causes them to sag off their men, and that half-yard of space leads to easy buckets. This is all without Dirk ever touching the ball. Leave alone when he gets the ball, the double comes, and the entire floor gets distorted for the defense.

If you want, I'll dig up video and show you dozens of individual plays where Dirk does this.

THAT'S the offensive impact Dirk has. The threat of scoring he presents is far more dangerous than the actual scoring he does.

KG may be a better passer skill-wise but the assist numbers are a little hollow to me, because it's not like he was ever drawing doubles and getting his team-mates open shots. Why? Because he was never the threat Dirk was.

KG's passing is a non-factor to me in this comparison. The only reason his assist numbers are higher than Dirk's is because the Mavs are so damn good at swinging the ball around that Dirk just got a lot of hockey assists. Dirk's opportunities created FAR outnumber KG's.

So really, when you look at it holistically, your argument that KG does "all the other things on offense better" is wrong. It's not a question of individual skills but the whole package. And because of Dirk's gargantuan gap in scoring, he is able to use his skills to the betterment of his team far more. It's all well for KG to be a good passer, but it's not helping his team a third as much as Dirk if he's not drawing attention to open up the floor for his teammates.

So I disagree that you think scoring is only 10% of the game: when you're as good at it as Dirk, it's far more, because it's not just the scoring but the THREAT of scoring that makes a huge difference.

I'll go with you to the extent that my 10% statement is an oversimplification, and hence effectively exaggeration.

The reality is though that when we look at the regression data, Dirk's offensive edge over KG isn't so, so huge, and then there's the whole defensive side to consider. When KG comes out ahead, it should hardly be a shock.

[colts18]

NPI RAPM:

98: DRob +3.77, KG +2.55
99: DRob +6.13 KG +2.31
00: DRob +3.59, +3.16
01: DRob +4.1, KG +1.5
02: DRob +3.0, KG +2.8


Robinson beat him out every year from 98-02.

[ardee]

Why do you think individual ortg describes offensive value better than for example ORAPM?

Also, you still didn't show that Dirk's offense > KG's defense and without that it's impossible to say that Dirk was as valuable (or more) as Garnett.

Because RAPM has shown ridiculous results I have pointed to time and time again. I'm not turning this into another RAPM debate, they're ruining the threads. Safe to say I don't use that stat at all, and never will.

ORtg + USG is the best measure of individual offensive production.

Dirk's offense vs KG's defense?

Ok. Dirk has multiple times taken a very average to below average supporting cast to phenomenal team offenses.

Let's look at 2006. The starting 5 were: Devin Harris, Jason Terry, Josh Howard, Dirk, and Eric Dampier. That unit was the starting 5 to the team with the best offense in the league, and went to the finals.

Remove Dirk and put in a replacement level PF. How is that team not the worst in the league on offense? JET, Howard, none of them were reliable shot creators. It was Dirk's defense-distortion that opened up all kinds of lanes and shots for them.

Dirk NEVER had another reliable shot creator on his team. Yet, here are the offensive results for his teams:

Dirks teams:
2001: 4th best Offense
2002: Best Offense in the League
2003: Best Offense in the League
2004: Best Offense in the League
2005: 4th best
2006: Best Offense in the League
2007: 2nd best
2008: 8th best
2009: 5th best
2010: 10th best
2011: 8th best

Now, meanwhile, let's take a look at what KG had to work with defensively:

I'm not saying it was anything great, but it's probably on the same level, or not too much worse defensively, than what Dirk had offensively.

Rasho was a solid center defensively, helped co-anchor some epic Spurs defenses with Timmy in the mid 00s. I don't know or care what his RAPM was, but his defensive On/Off from '05 and '06 was -6. He was definitely capable of making a good defensive impact.

Billups we all know was one of the premier defensive point guards in the league in Detroit.

Trenton Hassell, a real defensive specialist.

Sam Cassell, who colts18 earlier claimed was a better defensive player than early 2000s Kobe.

Other serviceable if not solid like Blount and Peeler.

I'm not saying the defenders KG had over his career were amazing, but it's hardly as if he was dealing with Smush/Kwame or Larry Hughes/Big Z. If anything, they weren't that much worse than what Dirk had on offense. Yet, look at the defensive results that Minny had in KG's prime:

2001: 16th in the league
2002: 15th in the league
2003: 16th in the league
2004: 6th in the league (the one year the results actually showed. I think this was the Boston syndrome: the presence of two other creators in Cassell and Sprewell allowed KG to exert more energy defensively)
2005: 15th in the league
2006: 10th in the league (not bad, 28th in offense though)
2007: 21st in the league

Even if you really want to slander his defensive supporting cast, it still doesn't explain why Minny's defensive results were SO much worse than Dallas' offensive results when Dirk had a pretty bad offensive supporting cast as well.

We're talking a consistent top 5 offense vs a consistent average to below average defense.

The only explanation is Minny KG isn't as good defensively as you think.

Which leads me to the general problem with KG here.

Minny KG and Boston KG are two different players. People are combining his offense at Minny with his defense at Boston while arguing with him. He was NEVER that impactful on defense in Minny, he spent too much energy on offense.

It doesn't work that way. If I took Kobe's hops in 2001, his three-point shooting in 2003, with his smarts in 2010, I'd have the arguable GOAT offensive player. I can't do that obviously.

So yeah, I think results speak for themselves. Dirk offensively > KG defensively.

You can argue a small difference in results but not when the gap is so big and consistent.

[Colbinii]

A big thing you guys need to consider with KG/Dirk is not only was KG the main component and anchor on defense, he was also the main component and focal point on offense. Dirk didn't have to do both, he got to focus on one side of the ball, where KG had to be "The Man" on both sides of the ball.

[DQuinn1575]

Why are people arguing Garnett versus Robinson, Dirk, or Duncan?

Have Dirk, Robinson or Duncan been voting on at 8?

Compare Garnett to Bird, Magic, or Hakeem - I think those are the 4 guys with votes right now.

Or ask what do you think about so and so - Walton, Mikan, K Malone, Durant, Oscar, whomever - if you think there might be a case for them at 8.

I respect everyone's opinions (probably the ones that agree with me more - but I don't want to read about someone's opinion for the guy they think is #11 when we are discussing 8.

And don't compare someone's #8 guy to your #12 guy.

[Purch]

Those Minnyy defensive numbers are absolutly underwhelming. Especially if people try to make the case that KG's impact on Boston's defense was greater than Tom Thibs (who has produced elite defenses every season the past 4 years with significantly worse defensive personal, and whiles giving minutes to a defensive liability like Boozer) The narrative people create about KG being the driving force behind Boston's defense doesn't seem nearly as compelling or as true as Thibs defensive schemes producing an all time elite defense, when having the personel to execute it at it's fullest.

It's like Garnett had 8 years of leading Wolves teams to average or below average defensive ratings, joins up with an all time elite defensive coach in Thibs and produces an all time elite defense. Yet, the only conclusion people draw from that, is that KG's defensive abilities simply shined through by not having to carry the scoring load. Even though, Thibs has produced elite defenses without a defensive anchor anywhere near KG's level(or so his fans would claim). It seems pretty clear the main reason that team was so great defensively.

For a guy people claim to be on the level of an anchor like Hakeem, leading a team to a 21st ranked defense is unacceptable, Especilly if people argue that his prime lasted till 2008.

[shutupandjam]

Why are people arguing Garnett versus Robinson, Dirk, or Duncan?

Since a few people are arguing in favor of Garnett, others are countering that argument by essentially saying "Garnett can't be in yet, he's no better than Robinson/Dirk." Garnett supporters are arguing "Duncan was voted in at #5 and Garnett is as good as him." So it has less to do with those other guys and more to do with whether we should be voting for Garnett at this point.

[acrossthecourt]

So here's the problem with people wanting to dock Garnett because of RAPM:
1) By RS numbers, David Robinson is top five ever, depending on how you deal with longevity. But the problem is the post-season where he was embarrassed in his prime. Robinson's post-season failures are more pronounced than Garnett's. Someone broke down Garnett's post-season "failures" and they're not as bad as people think. Just because a guy is fairly or unfairly known for not scoring well in the playoffs doesn't mean he's as bad as Robinson.
2) People say if Garnett is top ten by RAPM, then why don't you choose Dirk first? But Garnett's numbers are better. I don't see what's going on here because Garnett clearly has an edge. I think people want Dirk to be top 15, and the Garnett backers agree. There was a large push for this last season, and that was before Dirk put up yet another quality season while Garnett barely a weird year in Brooklyn. Garnett's older and has more seasons under his belt. That was another argument for him over Dirk.

[ardee]

Those Minnyy defensive numbers are absolutly underwhelming. Especially if people try to make the case that KG's impact on Boston's defense was greater than Tom Thibs (who has produced elite defenses every season the past 4 years with significantly worse defensive personal, and whiles giving minutes to a defensive liability like Boozer) The narrative people create about KG being the driving force behind Boston's defense doesn't seem nearly as compelling or as true as Thibs defensive schemes producing an all time elite defense, when having the personel to execute it at it's fullest.

It's like Garnett had 8 years of leading Wolves teams to average or below average defensive ratings, joins up with an all time elite defensive coach in Thibs and produces an all time elite defense. Yet, the only conclusion people draw from that, is that KG's defensive abilities simply shined through by not having to carry the scoring load. Even though, Thibs has produced elite defenses without a defensive anchor anywhere near KG's level(or so his fans would claim). It seems pretty clear the main reason that team was so great defensively.

For a guy people claim to be on the level of an anchor like Hakeem, leading a team to a 21st ranked defense is unacceptable, Especilly if people argue that his prime lasted till 2008.

Yeah, if someone is THAT good I see no reason why the results can't at least be passable.

Look at Robinson or Hakeem's casts... Nothing to write home about either, yet they regularly anchored defenses among the best in the league.