RealGM Top 100 List #11

[colts18]

i should double-check and finish the numbers, but I looked at how WS/48 changed with ABA players and NBA players from 1976 to 1977. Both sets of players declined, but the NBA declined more. Given their records in those exhibition games between leagues and more data about players switching teams, it appears the ABA was at lest as strong as the NBA in 1977, though not as big, and got progressively worse as you went backyards. But for like three/four years they were comparable.

Here is a post I made about year ago about the ABA. The top guys in the ABA did well in the NBA but the NBA had more depth

The ABA was a fine league with its top players. If you look at the 77 season, 6 out of the top 11, and 10 of the top 20 in PER had ABA roots. But the NBA beat out the ABA in the next tier of guys. For guys 21-40 in PER, only 5 were ABA guys (6 if you count Rick Barry).

[Clyde Frazier]

How about Moses?

1981 Rockets
40-42, Moses led sub-.500 team to the NBA finals

1982 Rockets
46-36, Moses' last season with Rockets, made playoffs

1983 Rockets
14-68, one of the 10 worst teams in history, won 32 fewer games because Moses left

No, they would not have made the playoffs with Moses either ... definitely not. And there was a lot of change on that team. How many wins/SRS did Moses add to the 76ers in 1983?

RS: 7 wins.
PS: 6 fewer losses, 2 more wins, 1 more championship

Single season changes like that are noisy and depend on a lot of other variables. It's interesting to look at to gauge impact, but it's not close to being precise. And yeah, Moses has a few lukewarm team win changes when being moved.

This was a move during Moses peak. Those other moves were non-peak team changes.

Hate to nitpick, but do you really need to do the red font as opposed to just bolding + italicizing what you're referencing? I keep thinking i'm reading moderator interventions or something.

[acrossthecourt]

Someone may work on that this year. There is the scoring margin adjustment, but that's a little different. And Garnett was on some very good teams for a while too and still had sky-high +/-.

Remember, as I cited in my tome, Boston *far* exceeded expectations. Garnett's more useful for better teams, especially if you look at his skillset.

Right, but the problem with 08 is that the Celtics had Thibs as well, a guy who would go on to make some remarkable defensive impact on the Chicago team for whom he is now the head coach, yes? It makes the full measure of KG's impact hard to properly assess, though obviously his injuries do help correlate his presence with maximum defensive efficacy. His impact on D was obviously great, but it was obviously more than just him, so that means he can't take full credit, IMO. It's something to think about, I believe, worth consideration.

I wonder, as always, are people conflating greatness and individual player ability, which are related, but not exactly the same thing. KG was clearly a remarkable player: a very good overall offensive player (second-tier offensive star) and an A-list defensive presence who did some great things and was probably miscast as a first-option offensive star due to the bungling inadequacy of Minnesota's management team. As ever, I also question the value of single-number metrics as a Grail stat used to justify serious change in a player's ranking. KG was obviously very good across a variety of analytical POVs, but his case seems entirely made on the basis of RAPM and that is not a comfortable thing for me.

Boston had a remarkable defense, but it behooves people to remember that in 2008, they added KG, Thibs had a wicked and keen scheme, we aren't talking enough about James Posey, or the coaching impact of KG playing more limited minutes as a way to keep him fresh, right? Pierce and Ray were both playing basically the best defense of their careers even when Garnett wasn't on the floor, in part because of reduced offensive load, in part due to coaching... in part due to intangible leadership elements from KG (he's a great communicator on D, surely).

I see the "lightning strikes 14 times" thing and it's great, but what makes me stop and pause is that KG's impact is supposed to be so remarkable, but we still haven't resolved whether that should matter in the absence of team results or not. This is about greatness of career, right? Not about who took crappy teams to moderate success, so theoretically there should be some element of winning bias inherent to the very concept we're discussing, and that should have some weight as far as KG's career is concerned. If not that, then his inability to fill the totality of the role which he was assigned in ways which others of his peers were able to manage. And again, we're looking primarily at RS RAPM, if I'm understanding this correctly, when most of the criticisms of Garnett pertain to his playoff offense, which is considerably worse than his regular-season offense, yes? There's a gargantuan difference in his performance on that end of the floor once the playoffs begin, so I'm given to wonder just how much that affects an impact stat like RAPM. Perhaps there's data at which I've not looked?

These are the questions floating through my mind, since I"m admittedly not really deep into the +/- stuff.

I wrote a few pages about how Garnett's special and why those teams failed in Minnesota and I did this *without* mentioning RAPM in the body, so I'm tired of people complaining about +/- being the only thing that supports Garnett. It's not.

I've responded to most of these points earlier, and Posey? He's not a game-changer at all and an overrated defender.... Boston had their second best season on defense without Thibs, by the way, in 2011. Garnett had him for a short while, but coaching is what's setting him back in the argument? Duncan played his entire career with Popovich, Kobe had Phil Jackson for a long time, etc....

We still haven't resolved the team issue? I"m not going to rank guys based on how good their teammates were. That's how I'm looking at it. Even then, Garnett had a great career. MVP, a title, was really close to another title in 2010, and arguably should have won at least one more MVP. West only has one title. Kobe has two without Shaq. Dr. J had one in the NBA. Karl and Barkley have zero. We're not comparing Garnett to Duncan anymore.

Bolded part: I know I've mentioned several times that most RAPM models use playoff games. And on page nine I looked at several seasons of KG"s offense in the RS versus PS and found *no* statistical significant drop.

[therealbig3]

KG vs Kobe: a multi-post analysis...Part 4

Part 1
Part 2
Part 3

I've covered the non-box score aspects from 3 different perspectives (with/without, on/off, and RAPM), and all 3 perspectives say that KG was the better player throughout his career. So now I'm going to look at the conventional method of evaluation: the box score.

KG vs Kobe, box score

Spoiler:

One thing I'm immediately going to say is that players that impact the game primarily through their offense always tend to look better via box score than players that impact the game primarily through their defense, so this is why it's hard to just compare stat lines and conclude that the one with the flashier stat line is better (and I don't need stat lines to tell me that Kobe is a better offensive player than KG anyway). The box score can do a pretty good job (sometimes) of gauging offensive impact, but it does an absolutely terrible job of gauging defensive impact, so Kobe is obviously going to be favored in a box score comparison with KG.

And in fact, that's not even completely true...KG's production in the RS actually trumps Kobe's. It's the PS where there's this huge perceived gap, and that's what I'll look at.

But when you see KG's playoff career, I agree that the fact that his 99-03 playoff career consists purely of 3-6 playoff games per year contributes to a ton of variance. And it's also not fair (to Kobe) to compare 99-08 KG's playoff production to Kobe's 01-10 playoff production, because Kobe played WAY more playoff games.

However, Kobe himself doesn't look like that great of an offensive player in the playoffs in terms of efficiency before 06 anyway. And everyone generally agrees that his best stretch as a PS performer was 06-10. And that gives a much more comparable sample of playoff games (06-10 Kobe played 79 games, 99-08 KG played 65 games). In fact, if we only include 08-10 Kobe, which everyone points to as his most complete and most dominant stretch of playoff performance, we have 67 games for Kobe...almost an identical sample size.

99-08 KG in the PS per 36 (65 games): 20/11/4 on 52% TS and a 106 ORating

08-10 Kobe in the PS per 36 (67 games): 26/5/5 on 57% TS and a 115 ORating

You take into account defensive environment (99-04 was a much harsher defensive era than 05 and beyond), as well as the difference in support cast (08-10 Kobe was on a great team, while 99-03 KG was not...possibly 04 as well if you consider Cassell's injury), and that 5% difference in TS and that 9 point difference in ORating gets a lot closer.

So what we see is that there actually isn't a huge difference in efficiency between prime KG in the playoffs and the best we've seen from Kobe in the playoffs. And furthermore, this kind of gap in terms of box score production is pretty much what we always see when you compare a star perimeter player to a star big man whose primary value is on defense. Do the same comparison with Kobe and Duncan and you'll see Kobe with an efficiency edge there as well, as well as a flashier stat line.

Conclusions: KG vs Kobe

Spoiler:

I've looked at KG vs Kobe over the course of 4 posts, and it seems pretty clear to me that all the statistical evidence points to KG as the better player. Clearly surpasses Kobe from any +/- variation of stats you want to use, and in terms of box score production in the playoffs...Kobe's advantage has been comically overstated, when you consider the inherent advantage he would have to rack up stats as a star guard, as well as the difference in team context and defensive environment. When you compare Kobe pre-06 in the playoffs to Garnett pre-06...their offensive efficiency actually looks remarkably similar.

I just can't justify Kobe over KG from any statistical perspective. The slight advantage that Kobe has over KG as from a box score perspective is pretty common when comparing a star perimeter player to a star big man, and KG consistently trumps Kobe from a +/- perspective, which tells me that it's pretty clear that KG's defense more than makes up for whatever offensive advantage that Kobe has (which is reflected in the box score).

[Doctor MJ]

How well this is reflected in RAPM data is a bit fudged, as there's limited available for Malone's career. His RAPM in '98 thru '00 look pretty good, though......"generally competitive" with the league's elite (especially in '98).

With the exception of '04, the vast majority of KG's hyper-elite RAPM seasons came in years where he was playing reduced minutes (<33 mpg every year since '07, and even <30 three times). Doctor MJ, probably the biggest proponent of RAPM data here, has made comment---months ago in a debate about John Stockton, and his near-stellar RAPM stats during the twilight of his career---about how impact stats often improve with reduced minutes. So presumably the same applies to KG.
Malone's RAPM in the late 90's is comparable to KG's RAPM stats for all years he was playing big "star level" minutes (again, with the exception of '04, when KG was insanely good).

Hmm. Let's see. I'll start with the stuff clearest in mind, the KG stuff.

Garnett has 2 years with a scaled RAPM north of 11. They came in years where he played 38+ MPG in Minny.
Garnett has 7 years with a scaled RAPM north of 9, 4 of them 38+ MPG in Minny.

Now, by what I have seen in the data, Duncan is the 4th most impressive guy we've seen over a long period of time. Duncan has one season where he played 38 MPG and edged past 9 (9.16), and another season where he edged past 8 (8.12).

So, in terms of hyper elite RAPM at big minutes, Garnett utterly dwarfs anything we've seen from guys except Shaq and LeBron. This plays into why I conclude that no matter how you look at it among players we saw plenty of in this data-driven era, there are 3 guys in the top tier, and then a drop off.

Getting over to Karl Malone, our earliest PI RAPM does indeed make him look quite good. He had a +9.01 in '98 at age 34. He's good, in RAPM, or any other way you look at it. If he were to take the 11 spot, it would make plenty of sense to me.

At the same time, that's basically the 11th year of Malone's superstar run (I think starting from Malone's 3rd year, which happens to be what the data says for Garnett makes sense, feel free to argue otherwise). So even if we accept that good-sized superstar prime and credit him with more years beyond that, he's still only approaching Garnett's 15 years of superstar impact. You can still give Malone the longevity credit based on more minutes played, but we're not talking about some big edge here.

And then there's the matter that if we're talking RAPM, things drop off after '98. He falls to +5 levels the next year even prior-informed, and by '02 he doesn't look anything like a star. The tale of Malone, according to +/-, is that he's one of those guys who stayed on trying to be his peak self even as it stopped having anywhere near the same impact, and hence from a +/- perspective, I wouldn't call Malone a longevity GOAT. Garnett is, Stockton is, Malone is not.

[FJS]

There's a lot of KG love here....

Altough he is a great player, he shouldn't be mentioned yet (in my opinion, of course).
He has played a lot of years, but his prime is not bigger than 10 years.
The only player mentioned right now who has missed playoffs 3 years in a row (or even not in a row) in his prime.

Don't forget Oscar Robertson

And this is why i'm not considering oscar as high.

[tsherkin]

I wrote a few pages about how Garnett's special and why those teams failed in Minnesota and I did this *without* mentioning RAPM in the body, so I'm tired of people complaining about +/- being the only thing that supports Garnett. It's not.

I agree. I've been in the pro-Garnett camp on more than one occasion when people have said ridiculous things about him. It's very clear that he was a fantastic defender, and he was indeed a very good regular-season player on offense. Clearly in Boston, he was the anchor of their remarkable defense, and a well-deserved winner of the DPOY. He's a player of sufficient caliber to belong in this discussion, so it's key to remember that while people may or may not agree that he belongs at a given spot, he does at least deserve to be discussed (especially when you consider the varying criteria we all use to establish what we believe to be meaningful indications of "greatness"). It's not really a diss to the player to critique their career; it is, in fact, the point of this whole project to expose ourselves to differing opinions.

I've responded to most of these points earlier, and Posey? He's not a game-changer at all and an overrated defender.... Boston had their second best season on defense without Thibs, by the way, in 2011. Garnett had him for a short while, but coaching is what's setting him back in the argument? Duncan played his entire career with Popovich, Kobe had Phil Jackson for a long time, etc....

Posey wasn't a game-changer, nor am I trying to say he was a huge thing, but the 08 team had a really good overall cast. They kept it working within the same system which Thibs established even after he left because they had good core continuity and team focus. And of course Rondo wasn't invisible in all of this, either, though he was less important earlier on.

I don't think coaching is a single element holding him back, no. You need a good to great coach to win titles, that's sort of implicit. Someone to change things up and get the pieces moving in the right way, in the same way that Phil did for Chicago or Pops did for the Spurs, etc, etc. Don't go too far to an extreme with this stuff, we're talking about multiple variables being involved, not one dominating.

We still haven't resolved the team issue? I"m not going to rank guys based on how good their teammates were. That's how I'm looking at it. Even then, Garnett had a great career. MVP, a title, was really close to another title in 2010, and arguably should have won at least one more MVP. West only has one title. Kobe has two without Shaq. Dr. J had one in the NBA. Karl and Barkley have zero. We're not comparing Garnett to Duncan anymore.

Right, but when you're comparing him to a series of other players who are all theoretically on a similar plane in terms of ability, then individual accolades and team success do have to matter to one degree or another, and that's not an issue which has been meaningfully discussed.

Bolded part: I know I've mentioned several times that most RAPM models use playoff games. And on page nine I looked at several seasons of KG"s offense in the RS versus PS and found *no* statistical significant drop.

That makes me question the stat even more, though, because he's demonstrably worse on offense in the playoffs, and irrefutably so. If it doesn't pick up on that for him as a high-usage volume scorer in Minny, then what's it actually saying?

[Texas Chuck]

I've covered the non-box score aspects from 3 different perspectives (with/without, on/off, and RAPM), and all 3 perspectives say that KG was the better player throughout his career. So now I'm going to look at the conventional method of evaluation: the box score.

Can someone help me out with the degree to which these 3 perspectives differ? It seems to me they are all kinda using a very similar persepective, but maybe there is a real significance in some of the nuance that Im missing.

[Texas Chuck]

[

Hmm. Let's see. I'll start with the stuff clearest in mind, the KG stuff.

Garnett has 2 years with a scaled RAPM north of 11. They came in years where he played 38+ MPG in Minny.
Garnett has 7 years with a scaled RAPM north of 9, 4 of them 38+ MPG in Minny.

Now, by what I have seen in the data, Duncan is the 4th most impressive guy we've seen over a long period of time. Duncan has one season where he played 38 MPG and edged past 9 (9.16), and another season where he edged past 8 (8.12).

So, in terms of hyper elite RAPM at big minutes, Garnett utterly dwarfs anything we've seen from guys except Shaq and LeBron. This plays into why I conclude that no matter how you look at it among players we saw plenty of in this data-driven era, there are 3 guys in the top tier, and then a drop off.

Getting over to Karl Malone, our earliest PI RAPM does indeed make him look quite good. He had a +9.01 in '98 at age 34. He's good, in RAPM, or any other way you look at it. If he were to take the 11 spot, it would make plenty of sense to me.

At the same time, that's basically the 11th year of Malone's superstar run (I think starting from Malone's 3rd year, which happens to be what the data says for Garnett makes sense, feel free to argue otherwise). So even if we accept that good-sized superstar prime and credit him with more years beyond that, he's still only approaching Garnett's 15 years of superstar impact. You can still give Malone the longevity credit based on more minutes played, but we're not talking about some big edge here.

And then there's the matter that if we're talking RAPM, things drop off after '98. He falls to +5 levels the next year even prior-informed, and by '02 he doesn't look anything like a star. The tale of Malone, according to +/-, is that he's one of those guys who stayed on trying to be his peak self even as it stopped having anywhere near the same impact, and hence from a +/- perspective, I wouldn't call Malone a longevity GOAT. Garnett is, Stockton is, Malone is not.

I can't at all get behind the idea that KG has more longevity than Malone. At all. I have tried to stay out the whole RAPM debates, but its becoming more and more clear that its casting this huge shadow over any and all discussions involving KG to the point of making much discussion useless. By all reasonable measurements Malone is showing to have great longevity.

[drza]

Spoiler:

I've seen some people say that the onus is on users of RAPM to justify its use as a stat. Perhaps, to an extent. However, as fpiii points out, the approach is peer reviewed and proven in other applications. And repeated measures strongly supports validity. So in this instance, the question becomes one of: of how could one particular player be consistently correlated with major positive team scoring margin results across such a wide variety of circumstances and so many years, and it all be explained away as a fluke or an unsupported throwaway "it's a line-up effect" or "well, there are sometimes strange sounding results in single studies so I disregard this"? At some point, doesn't the odds of lightening striking 14 times in the same spot start to suggest that maybe this effect is real?

No offence drza, but I really don't have an appetite for RAPM debates anymore because they just don't go anywhere. There is no "line-up effect"....the data itself is based on lineup data. The composite +/- at the very heart of each possession represents a lineup...not an individual. You can't use RR on that data(even large samples) and expect it to magically derive individual impact, it's simply not possible. RR is extremely useful for regularization, but the problem you're trying to solve requires the data to first be individualized to an extent. If that happened, then yes RR would be a sound method of handling that data. hence why I've mentioned Synergy type data. You may find some trends that were influenced heavily on the rotations Player X was in, but to surmise the result to the individual is incorrect.

I understand about the RAPM debates, but this is actually an important point and worth getting a level of understanding. And you'll note that you've been espousing this "only get lineup results" for a long time now and this is the first time I've engaged you on it.

The problem is, you're factually incorrect.

And not about how to interpret KG's RAPM vs Kobe. You're factually incorrect in how you're describing regression results. This isn't a matter of opinion. Regression, literally, is a technique that takes in a bunch of inputs and results, and identifies which INDIVIDUAL(s) in that group of inputs are most correlated with the given outputs. Again, this isn't something to have a take on. This isn't something to debate. Factually, this is what regression does.

In the case of RAPM, the inputs are all of these different line-ups that you keep mentioning and the outputs are the scoring margins of these line-ups. There are tens of thousands of different line-up combinations, and a corresponding number of different scoring margin outcomes. The regression part of RAPM, literally, is identifying which individuals within those many line-ups are most correlated to given scoring margins. Point-blank period. The regression part of RAPM LITERALLY is doing what you're saying is "magical" or it can't do.

Now. The issue with the regression part of RAPM is that this is an ill-posed problems. If we scale it back and think about high school algebra, it's something like having more equations than you have unknowns. So the regression can't yield an exact number for each player, but instead can identify an answer range within which the correct value should be found.

This is where you start hearing folks talk about "priors", and where the "ridge" part of the ridge regression comes in. By using some amount of previous data to give a reasonable starting point idea for players (instead of starting from a random value), it helps make that "range where the likely value will be" smaller. You often see people say results are "noisy"...this is what they mean. The amount of room in that range is the amount that the regressed value could fluctuate from correct, and that is considered "noise". So using priors and lowering that noise is a good thing. Then, the "ridge" part of the analysis helps to avoid giving players that don't have enough representation in the line-ups too undue of an influence and also helps a bit with the issue of collinearity. All of these things, again, work to reduce the noise and help make the regression better. Not perfect, but better.

But again, the point is, it's important to realize that what you're saying about regression is factually incorrect. It's a particular problem because you are eloquent, you post frequently, and the things that you're saying sound just close enough to reasonable that others that aren't familiar with how RAPM works might believe you. Especially if, like you, they don't like what those RAPM results might mean. You don't have to love or use RAPM. That's your prerogative. But you shouldn't continue to press forward with a claim that literally isn't correct.

Also, in regards to lightning striking, the answer is not complicated. Player X simply sustain similar roles throughout his career, and was utilized in rotations that functioned the same. That's not unusual for basketball. RAPm results aren't strange, or "noise" or unreasonable...they're just misinterpreted as reflections of the individual instead of lineups.

As for this part, the underlined is CLEARLY not the case for Garnett. I wouldn't even think that needed to be pointed out. He's been a SF, a PF, and a center. He's played with casts varying from terrible to below average to average to above average to great. He's been a primary scorer, a secondary scorer and a tertiary scorer. He's been the lead distributor and primarily a finsher. He's played in unipolar offenses and ensembles. He's anchored defenses from the top of the key and from the paint. He's played for at least 6 different coaches, on multiple teams. Your answer for Garnett's results can't possibly be that he's been in a similar role, with similar rotations and utilizations. It just can't be.

[therealbig3]

That's the prevailing notion, and it's one that I generally agree with...but that's the funny thing about Garnett. After he goes to a good team in Boston, his RAPM remains super-elite. Even in his post-prime years.

Keep in mind though... MASSIVELY reduced offensive load, fewer minutes and great coaching. Pops has shown us that it's quite possible to maintain the productivity and utility of youth as long as you limit a player's minutes per game. Garnett exerted a very obvious and palpable force for those Boston teams, but he was also doing it less and less each year, so this gets back to his early days with Minny when he was still a teenager and posting those ridiculous values as well, even when his raw averages and such seemed to imply that he wasn't that consequential a player.

Well, the furthest back RAPM goes as far as I know is 97, which was Garnett's first AS season, so to me, if he has high RAPM values in those years (haven't looked into acrossthecourt's numbers yet), it jives with what people were seeing back then with him.

As for his reduced offensive role in Boston, that's true, but IIRC, we see a clear shift from Garnett's impact on offense to just concentrating way more on defense. And if he had a reduced offensive role in Boston, he clearly had an increased defensive role in Boston. Impact is impact, and the fact that Garnett was maintaining the same level of impact regardless of the team he's on simply by changing what area of the court he focuses most of his energy on, that's pretty incredible.

As for the Popovich example, we actually haven't seen that happen with Tim Duncan. Duncan was one of the elite superstars in terms of RAPM during his prime...and once he started slipping out of his prime, he's still been strong in terms of his values, but nowhere close to what he used to be...even though his box score stats look similar and even though he's had a career resurgence since 2012. In his case, RAPM actually does a good job of tracing his career path as an individual player, despite Popovich's adjustments. So the fact that even as he aged, Garnett was still the lynchpin of some really good Celtics defenses (while still being a contributing factor on offense), and that bears out in RAPM, tells me something about KG. Honestly, with the exception of 09 and 10 (injury-riddled years), there was never a doubt in my mind who the best player on the Celtics was since 08...it's always been Kevin Garnett.

[HeartBreakKid]

Some recent post about West have made me consider taking West over him, but for now my brain still says Big O, mainly based on eye test. Watching Oscar Robertson play was pretty amazing, he looked every bit as dominant as any modern player, and I only watched him during his Bucks days. His mid range game and post game were excellent, and he had the type of body that would make him an elite athlete in the modern era as well.

It's hard for me to ignore him and West's awesome efficiency during a time when every shot well below the curve. That type of mid-range dominance and slashing ability is pretty impressive, even if the defenses they went against may have been inferior, they faced very different problems from modern players, primarily the clogging of the paint, which meant Big O was either an amazing mid range player or amazing at getting to the rim, from what I've seen it was both.

His passing ability I think goes without saying, he's as good as a passing guard as most PGs, probably better than most score first PGs I would say. His off/on offensive impact is quite impressive with the Royals, to be honest watching Big O, there was really nothing that made me think he was any inferior to Magic Johnson, and I have Magic 10th on my all time list.



His biggest competitor to me right now is probably Garnett. A lot of the "if Big O was so great how come his team didn't do this or that" kinda applies to Kevin Garnett as well. Garnett is easier to feel comfortable about because we have more stats and footage of him, but it's hard to deny that Robertson was probably the best offensive player up unto Bird came around, I suppose second best if West is more your fancy.



The team stats are nice evidence, but the primary reason why I don't rank Dr.J, West and Bryant over Oscar is really from the eye test. I could be convinced, more so of Dr.J and West, as I know quite a bit about Bryant already, but it doesnt seem like there are too many people pulling for Dr.J and West to be put over Robertson.


As for Bryant, what Big O gives up in scoring, I think he makes up with superior playmaking. His PNR play was really impressive considering the time he grew up in. His playoff run the year he won the title with Kareem was an awesome show. I think Big O could hurt an opponent in more ways offensively than Bryant could, which made him more portable and probably harder to game plan against. There's also the chance that Bryant may shoot himself out of games if defenses really key in on him, as he will tend to settle for bad shots instead of trusting his teammates or giving them better looks (which he is capable of, but his IQ and ego get the best of him I believe).


So my vote goes to Oscar Robertson.

Right now the closest player who will make me retract my vote is Kevin Garnett. So if anyone thinks KG is better than Big O, feel free to share. Keep in mind, that it is close enough that I would realistically change my vote. (like, maybe 55-45 in Oscar's favor right now).

[therealbig3]

I've covered the non-box score aspects from 3 different perspectives (with/without, on/off, and RAPM), and all 3 perspectives say that KG was the better player throughout his career. So now I'm going to look at the conventional method of evaluation: the box score.

Can someone help me out with the degree to which these 3 perspectives differ? It seems to me they are all kinda using a very similar persepective, but maybe there is a real significance in some of the nuance that Im missing.

Well, you're right in the sense that they're all trying to measure the same thing: how much does a player add to his team?

With/without and on/off both look at a player's ABSENCE from a lineup, but in two different ways...on/off looks at when KG is playing, but when he's on the bench vs on the court. This is probably the noisiest measure of impact, because a lot of factors are going on: coaches go with "emergency" lineups that are only intended to stay out there for a few minutes, coaches play certain players with the star and certain players without the star to see what works the best, small sample size of "off" minutes for the star, generally, etc. However, over the course of a season, the noise gets minimized, and you can get a pretty good idea of how much a team depended on their star player to be out there. Sometimes, it also tells you a good deal about the 2nd unit and how effective they really are.

Meanwhile, with/without looks at how the team plays when the star player is not available at all. This is more revealing than just on/off imo, because now the coach KNOWS his star player isn't available, so now there are actually set rotations and and lineup patterns that are independent of the star player, and we can get a good idea of how the team functions, with the star player as the only variable. ElGee has actually gone far more in-depth than me on this, because he also controls for games where certain teammates might not have played as well.

RAPM is a more reliable version of pure +/-, and it only measures how a team does with a certain player ON THE COURT. Has nothing to do with when the player goes to the bench. +/- is the scoring margin for the team when that specific player is on the court. So it's looking at how good a team plays WITH the player in question.

So yeah, they're all trying to tell us the same thing, but they do it in different ways.

On/off tells us how effective a team's rotations and lineups are when the player is on the bench compared to when the player is on the court.

With/without tells us how effective a team's rotations and lineups are when the player isn't available at all.

+/- tells us how effective a team's rotations and lineups are when the player is on the court...RAPM is the attempt to measure the value of an individual's contribution to that +/-.

And sample size is what cuts down on all the possible noise that comes with this data. So when one player is consistently having huge values across the board, year after year after year, it's safe to say that this player is pretty special in what he's providing for his teams. And that's why a lot of people like Kevin Garnett so much...because there ARE big sample sizes that keep saying the same thing: Kevin Garnett is a monster.

[Owly]

Spoiler:

Also re: Oscar.

I respect his game a lot, and don't think it's weird to mention him here - obviously I did - but I'm noting people aren't really talking about West on their short list. I'm guessing what that means is that people are putting Oscar & West in the same "bin", rating Oscar ahead, and thus not seeing much need to consider West until Oscar is in.

If this is the case for you I'd urge to reconsider because Oscar vs West is a very good debate, and I'll just be brief here regarding Oscar:

-Didn't Cincy's weak records make you question Oscar's impact?
-If so, what assuaged your fears?
-If it was in any way shaped by the WOWY numbers that seem to "match" Oscar's box score in being superstar-like, consider that the most complete data we have on that front (ElGee's numbers) make West look even better than Oscar.
-All while West played on a contender and had to sacrifice primacy for much of his career to an inferior players with a lot of redundancy.

As an Oscar advocate

Cincinatti's weak record hasn't really made me question Oscar impact or at least it hasn't led me to change it. Firstly to acknowledge my biases I'm more familiar with boxscore metrics and so use them more. In general I've seen that in seasons close to their stats apex, my top 3 (Jordan, Jabbar, Wilt) had losing seasons (and others have had great individual seasons with teams around that level; McGrady '03, Wade '09, M Malone, Garnett etc).

Insofar as the weak record caused a concern my position/opinion has always been that the Royals were a cheap, poorly run franchise.

- The WoWY numbers aren't a big factor. They help to confirm or make you take a second look. But the little bit that I did and what I've seen from Elgee(so far as I'm comprehending it, and trusting it's accuracy) Robertson's impact looks huge from rookie to '68 (assuming, given his boxscore similarity his impact was around the weighted average of his impacts during that time. I'd suggest that meant he built a sizable lead in value added over his first three (perhaps even four, given West's injury in '64 and slightly smaller WoWY impact). But maybe I'm not great at interpreting this stuff and am looking for pro-Robertson stuff.

- Was Baylor a redundancy or was someone who freed up West for easier shots? And was West ready for prime-time right away (i.e. isn't their a reason, a valid one for the first 3 seasons, why West was the 2nd banana)? Was West getting any less shots than Robertson? And weren't West's continued absences (a) somewhat artificially inflating Baylor's usage and (b) forcing LA to retain Baylor as insurance against such injuries.

I'm not absolutely confident to the answer to all these questions. But my interpretations tend towards a sympathy towards Robertson.

And then too there's the fact that Oscar played those extra minutes, which too me, are valuable.

A final addition would be that every critical ranking until Bill Simmons' called Robertson better (well except the Thompson one noted below, it was a bizzare numbers including accolades based affair with Heinsohn at 16.5, Mel Daniels at 26.5, McGinnis at 37, I basically tend to write that list off) . Robertson was always listed as better, occasionally first and has an average ranking that had him at the bottom end of the Jordan, Wilt, Russell, Magic, Kareem, Bird tier (newer players excluded because you don't know when to start counting their votes). Then came Shaq (if you exclude an outlier ranking 38th all-time in 1996) and next but a clear distance behind, comes West. And from the bits I've seen around the time it's always been Robertson, West sometimes mentioned as the minority choice. I don't think critics get it right all the time. But for all of them to have got it wrong, and boxscores to have done so too (career metrics are close, despite Oscar's role change and with Oscar playing more minutes and Oscar peaking higher) I'd need something compelling. Maybe I haven't looked closely enough (e.g. at ElGee's stuff) but I haven't seen it yet.

I don't think it's crazy to choose West. He's got arguments for him (playoffs, defense - which is hard to gauge but certainly some very strong reviews, maybe WoWY). But I'd lean Oscar.

Maybe I'm just an Oscar fan whose stuck in a long held perception of Oscar's superiority. But there are my reasons.

Here's what I'd written for Oscar earlier, WoWY stuff takes the lead because that seemed to be a big part of the discussion at the time so I'd looked at Robertson's largest (prime) absence.

Though open to persuasion, my vote at this time goes to Oscar Robertson

I'll start with my case earlier, though now, this isn't so much for him to be in the argument as it is, "he's my choice right now".


My reasons for him over Bird. Longevity - prime wise: (higher mpg; better in terms of boxscore off the bat, though team level impact would seem to favour Bird; but particularly because he didn't suffer a major injury so early), and then in his post prime, Oscar found a very nice niche and helped a team be incredible, whilst Bird struggled somewhat, both in terms of clear boxscore diminuation, but also in finding a role, plus his D after the injury, from what I've heard, kept getting worse (iirc McHale and an annoymous teammate, supposedly Jim Paxson, felt or said at times Bird was taking too many shots).

My reasons for him over Olajuwon. Well it's doubts about Olajuwon's O. Others have noted some problems with his With/Without numbers in that area, and his his TS% was closer to average than great, despite his fabled post moves. Plus whilst he could make a lightning quick move upon recieving the ball, I get the sense that he held on to the ball a little too much (may be too influenced by Simmons saying he did so).

Iirc someone said something about not being convinced about him being better than West, but I can't find the quote so won't address it fully unless it comes up again here. But I think his D might be underrated and his combination of offensive efficiency at individual and team level whilst carry many large burdens (shot creation, playmaking-passing, very large rebounding responsibility for a guard) really resonates with me (also West got injured a lot). I'm not fully convinced Magic (generally, and here, considered his superior) was better tbh.

I also like how Oscar did in all-star games back when the games were competitive both in terms of accolades and team record (including victories over teams with Chamberlain and Russell, 3x MVP, plus once perhaps assisting in Adrian Smith winning it, cf Tall Tales p199). It suggests to me, albeit on very limited evidence that he could perhaps have had even more impact with good teammates (and at least reassures me this wasn't just numbers on a bad team, though tbh once efficiency is factored in, there's limits to how much being on a bad team helps your stat line).

Admittedly I'm not super systematic with this and the "reasons" (for Robertson over players x and y) are more rationalizations (I don't get my order by explicitly comparing each candidate at each spot (the sheer number of comparisons would be daunting, I just do my ad hoc "who's better" calculations. Not entirely happy with by process and internal consistency but would have to dedicate more time than I have- and even then not all source material is available the same for all eras. Anyway I put the "reasons"/reasoning in as I think Elgee kinda wanted some, probably rightly so).

Good stuff.

Re: Baylor, free up West? Well I think in general when you've got a situation where a guy scores more using less shots, it's hard to fathom that the guy doing less damage with more waste deserves a lot of credit for it. Then there's the matter that West's the guard, and West really never showed a major falloff without Baylor in his prime.

I want to be clear: I would never argue "West made the Lakers better than Oscar's Royals despite that dead weight Baylor". Baylor certainly helped the team. However given that Baylor is known first and foremost for his scoring ability, and so is West, you would simply expect redundancy issues to occur which would make on/off numbers smaller than they'd be on other teams that have no one known for their volume scoring ability on the order of Baylor or West. For West to have such huge on/off numbers given this is shocking to me.

Re: First 3 seasons. Fine to use that against West. To me it's clearly a combo of West being not quite as ready out of college and him not being handed a team to run like Oscar was. It's a factor for me as well, but it's hard for me to see it as a deciding factor.

Re: Effect of West's injuries on Baylor's usage. That's a good thing to bring up. It's hard for me to really swallow though given that if you knew nothing of West and just look at Baylor's numbers, you'd just assume he was continuing to do his normal thing. If you had some sources that talked about Baylor being forced to change how he played because of West's injury that would be helpful.

Re: Oscar's minutes. A good thing to bring up. From my perspective there are a couple things that make me not that concerned about the minutes, though I'll acknowledge it's still a pro-Oscar thing:

1) How did the injuries affect West's role in the playoffs?

In the '60s, West missed time enough to drop his games played to less than 70 in the regular season 4 times.
3 of those 4 times he was fit for the playoffs, and dominant all through deep playoff runs, and in each of those years West was clearly having big impact for the bulk of the season in which he played and his team was north of the 30s-ish win total you typically needed to make the playoffs.

2) When considering longevity, one of the things I tend to do is include playoff minutes. Now, it's not Oscar's fault that he played less in the playoffs, but if you're going to use wear & tear as a factor in a comparison, which is what longevity means really in a comparison between two guys who aged well, clearly all minutes played contributed toward that.

Total minutes played:
Oscar ~ 47.5k minutes
West - 42.9k minutes

It's a factor sure, and if you're someone who has been really caring about longevity a lot in this project, by all means continue. Nothing wrong with that. For anyone though who has brushed longevity aside in places though, I'd submit this is a place where it's hard to use longevity as much beyond a tiebreaker.

Re: Oscar typically seen as better. In terms of GOAT rankings I agree, but consider the year-by-year details on the matter.

The last time Oscar was in the Top 2 of MVP voting was 1965. Jerry West by contrast finished 2nd in the MVP voting from 1966 on basically as a given until 1972. Injuries got in the way of that streak, and I'm not saying those are not relevant to the discussion, but that reliably 2nd place finish, does that sound like someone to you that people were seeing as clearly worse than Oscar?

So I would submit that when you see Oscar as the overall better choice, people were already looking at the factors you mentioned above, along with Oscar's extremely gaudy early '60s stats when things were inflated like crazy and the fact that he came in as the best college player basically ever. Oscar began this comparison with a huge lead, and it was just really tough for people to shake that even as they were regularly rating West over Oscar on a year to year basis once the Lakers gave West enough primacy.

But as I say, I respect someone's choice of Oscar over West. I think Oscar was amazing. The reason why I'm focusing on West here is that I just think it's a little too easy for people to just default to Oscar.

With the Baylor thing it depends what's being argued. Did he take defensive attention away from West (I suspect yes), did he create open shots for West (again I'd suggest probably yes). Was he better than West, no. Was his larger role sub-optimal after those first three seasons, probably. Was their a duplication of skills and therefore a redundancy, I don't know. It would depend whether there was sufficient shot creation skills amongst teammates that they could reduce Baylor's load (changing year on year), or on whether West could have shot more and Baylor less at an tradeoff that would have been positive for the team. I sense you think that since they didn't fall off, West could take on Baylor usage and so was under-utilised? But if so isn't that somewhat on West? There's a swings and roundabouts element to this argument (If Baylor wasn't giving West open shots, then wouldn't LA holding up largely okay be to be expected, or conversely if Baylor was so great and not duplicating an somewhat existing skillset, less efficiently, how come LA didn't get much worse without him).

Whether or not getting the keys to the team might of helped West, he didn't warrent them in '61. He should have started more (though his minutes were ulitmately fine, but as it transpired it unnecessarily p****d West off). Baylor at that point versus West is no contest. I don't mind hypotheticals but it's hard to see West getting the keys to any team at that time.

Baylor sans West is just a hypothesis. But looking at Basketball-Reference his top 5 scoring outings in '68, West's longest absence were all in games without West. So I think part of the impression that Baylor was taking West's shots in the late 60's might be illusory based on games in which Baylor had to (try to) take up the slack.

Regarding longevity. Criteria might differ here. Some are okay if you miss parts of the season if you get into shape for the playoffs, and obviously that's better for your team than playing through stuff and hurting yourself longer term. But you're not adding value and fwiw you're possibly costing HCA (the value of this varying depending on length of series etc). There's a degree of the hypothetical about it, but with Robertson I feel confident in a given game he'll be available and performing at or close to full strength. West was gutsy so I believe he'd try, but there are a few absences. Even if I was a longevity is only a tie-breaker guy, Oscar is at least I'd suggest close enough to warrent a look at it. And even with a system that penalizes Oscar for bad teammates he has a not insignificant advantage. I'm not someone who writes off longevity. Whilst I'm less consistent than I'd aspire to be my criteria is something like championship value added (not quite the title probability stuff which has a specific meaning on here and has a significant playoff weighting doesn't really mind missing RS, nor WARP or Win Shares type idea where each amount of goodness added is of the same value, even though adding say .200 WS or WARP per 48 gives you a much higher team ceiling than two guys giving .100) or value added with high end bonuses or something like that.

Re: MVP voting. Robertson has a clear edge in MVP shares. That 2nd place in MVP run from '66 goes
66: 2nd
67: No votes
68: No votes
69: No votes
70-72: 2nd

So to yes to me that does mean that over the sum of their careers Oscar was consistently called better since he was still getting All-NBA first teams during that span (West was 2nd team in '68 and '69). MVP voting from that era in particular had a particular best player on the best team flavour (hell Johnny Kerr did better than Wilt in '63). And still Oscar got some mild consideration through to '68. I'd take later model West though I'm not sure how much Oscar was just fitting into a role in Milwaukee and sublimating his game (it's hard to argue with the team level results, though off the top of my head I recall with/without probably isn't in his favour here). But that's 3-5 years (depending on whether you're talking about MVP consideration, West being better per minute, or West having a larger overall impact). Oscar had the same spell of clear advantage earlier in their careers and has the better numbers and hasn't missed as many games. Even if you go in for "Well, West would have been in the MVP running without injuries and injuries don't matter if you make the playoffs" that doesn't necessarily put him anything more than even for that spell '67-'68 in terms of who was considered better. So I don't think people thought Oscar was better every year, year on year. But I think they thought he built enough of a gap for himself by '68 (or '69) that despite what West did thereafter it wasn't closable. And that's what pretty much every GOAT list has said since.

Regarding this little snippet "along with Oscar's extremely gaudy early '60s stats when things were inflated like crazy". I've seen this reasoning before and I don't like it. Because West played in the same years and had the same chance to put up inflated, extremely gaudy stats. He didn't. It's particularly galling when used as Bill Simmons does manipulatively in terms of his numbers dropping in an expansion era. But West is celebrated for happening to peak later in the apparently weaker league (and not penalized for not putting up numbers when he had the chance to). Okay I'm arguing with someone who isn't here now. Back on topic, if it's just one player was elite first and so it's hard to switch order, fine I can see there's an argument there, I don't know whether its true. At the margins I can see a case his peak is slightly over celebrated because of boxscore inflation (though his team pace was on the slower side, and his best, MVP, got a book written about it year was considerably slower than the pace explosion peak - 109.4 fga per game in '61 and 107.7 in '62 when Wilt got 50, Oscar got his triple double etc; but only 99.1 fga in '64 AND Robertson's teams were inevitably slower than the league average AND the triple double stuff wasn't talked about until the 80s but people always seem to have had Robertson ahead). I just don't see that West did enough to close that gap.

Anyway as before I can see why people would go for West based on slightly different criteria than my own.

[Texas Chuck]

yeah your explanation confirms how I thought of each of them. They are all measuring the same thing. Sure there are some small differences and they are worth looking at individually, but to call them 3 distinct perspectives is stretching it. I'm going to choose to consider those 3 "categories" as 1.

[tsherkin]

As for his reduced offensive role in Boston, that's true, but IIRC, we see a clear shift from Garnett's impact on offense to just concentrating way more on defense.

Yes, that was kind of my point. He had guys with whom to share the load, so all of his effort went into D, where he'd already been a very good player. With fewer minutes and less offensive responsibility, it's not surprising that he'd be able to exert a greater impact on D, which would balance with his lesser role on O (which also benefited, since he was a considerably more efficient player in that reduced offensive role). His 118 ORTG in 2008 was a career-high, the 58.8% TS the second-best of his career. That boost to offensive value also helped his impact numbers, I imagine.

And if he had a reduced offensive role in Boston, he clearly had an increased defensive role in Boston. Impact is impact, and the fact that Garnett was maintaining the same level of impact regardless of the team he's on simply by changing what area of the court he focuses most of his energy on, that's pretty incredible.

I don't debate that he was a very good player in Boston. I was just looking at why he was able to maintain that impact in Boston versus Minny. It's easier to put forth greater effort in fewer minutes, and especially when you're exerting less effort on the other side of the court. It's implicit that he would have to be very good to begin with to even be discussing the specific level of impact which we are now examining.

As for the Popovich example, we actually haven't seen that happen with Tim Duncan. Duncan was one of the elite superstars in terms of RAPM during his prime...and once he started slipping out of his prime, he's still been strong in terms of his values, but nowhere close to what he used to be

I meant more specifically his per-minute productivity in terms of scoring and rebounding, and the corollary stats like TRB% and the like.

So the fact that even as he aged, Garnett was still the lynchpin of some really good Celtics defenses (while still being a contributing factor on offense),

I wouldn't play up the offense, given how bad they got on O over time and post-08 KG's general suck-assedness in the playoffs from 2010-2013, offensively speaking. On D, absolutely: in his absence, they weren't the same team at all. And again, diminishing minutes and games played are concerns as well. Not too much so, given the breadth of sample his Boston career provides, but still worth examining.

KG is obviously a very good player, and one who adapted well as far as impact over time and shifted role, but I question the specific value of these stats when they are just so strangely astronomical for him.

[therealbig3]

yeah your explanation confirms how I thought of each of them. They are all measuring the same thing. Sure there are some small differences and they are worth looking at individually, but to call them 3 distinct perspectives is stretching it. I'm going to choose to consider those 3 "categories" as 1.

It is 3 different perspectives though, because they're NOT measuring the same thing. They're all used for the same thing, but they measure it differently, which is why I consider them separate categories.

I feel like you're only choosing to consider these categories as all the same thing, because they happen to all agree with each other in KG's case. But there are many instances where on/off and with/without say two different things, even though they both look at when a player is off the court. And RAPM can say something different as well. Sometimes, a player that look great in terms of on/off has a poor RAPM, and vice versa.

The fact that they all point to the same conclusion with regards to KG just increases the likelihood that KG, and not numerous other factors, are actually the reason for it.

[tsherkin]

I can't at all get behind the idea that KG has more longevity than Malone. At all. I have tried to stay out the whole RAPM debates, but its becoming more and more clear that its casting this huge shadow over any and all discussions involving KG to the point of making much discussion useless. By all reasonable measurements Malone is showing to have great longevity.

So let's look at that.

98-02, right?

98: 27 ppg, 10.3 rpg, 3.9 apg, 37.4 mpg, 53.0% FG, 59.7% TS, 118 ORTG
99: 23.8 ppg, 9.4 rpg, 4.1 apg, 37.4 mpg, 49.3% FG, 57.7% TS, 112 ORTG
00: 25.5 ppg, 9.5 rpg, 3.7 apg, 35.9 mpg, 50.9% FG, 58.2% TS, 115 ORTG
01: 23.2 ppg, 8.3 rpg, 4.5 apg, 35.7 mpg, 49.8% FG, 57.2% TS, 112 ORTG
02: 22.4 ppg, 8.6 rpg, 4.3 apg, 38.0 mpg, 45.4% FG, 53.2% TS, 107 ORTG

So from a RS perspective, 02 actually appears to be the pivot year (and was his final AS season). Utah ranked 1st in team ORTG in 98 at 112.7. Then 3rd at 105.8 (lockout year), 6th at 107.3, 3rd at 107.6 and 10th at 105.6. There's a pretty clear drop-off in team offensive efficacy at that point, but the lockout affected things in 99 just as it did in 2012, so there's that to consider. Still, the team was obviously worse after. Now, Stockton was playing fewer minutes and wasn't really the same from 98 forward because of his knee. Hornacek retired after the 99-00 season and was playing progressively fewer minutes, Malone was decreasingly effective until finally tailing off completely in 02, Shandon Anderson fell off of a cliff, Bryon Russell was still effective but not to the same degree... There were various wrinkles contributing to that decline, including the increasingly nasty defensive environment leading up to the rules changes following the 03-04 season. Utah played at 87.0 to 90.3 possessions per game from 98-02. Look at his WS/48:

98: .259
99: .252
00: .249
01: .217
02: .155

The tail-off still seems to be happening a lot later than RAPM indicates. He led the league in OWS with 12.1 in 98, then had 10.7 in 00 (only his 5th season with double-digit OWS, FWIW). There are lots of counterpoints to the RAPM data there that make me wonder about the stat.

ITO box score productivity (PER)?

98: 27.9
99: 25.6
00: 27.1
01: 24.7
02: 21.1

Again, seems pretty strong for at least 2-3 seasons after RAPM indicates the drop-off. I think RAPM might be underselling him, to be honest. I don't recall his defense dropping off to any significant degree, if only because he was never a wicked "go everywhere" defender in the first place, more of a stay at home player focusing on man D and defensive rebounding.

Postseason, perhaps, might illuminate?

TS% / ORTG

98: 53.4% / 105
99: 49.2% / 100
00: 58.4% / 113
01: 48.4% / 97
02: 46.9% / 98

So another point in favor of 2000 being his last really awesome season, though in deference, he was at 49.8% and 50.1% TS in 96 and 97, with ORTGs of 105 in both of those seasons. 95 was the last time he hit or passed 110 ORTG before 2000, so that's really not where RAPM was drawing its indicators from. I don't know, this is another puzzler where RAPM seems to be saying something at odds with what a lot of other data is indicating. His raw ON/OFF ORTG was +10.8 in 2001 (+20.1 in the playoffs), then +4.0 in 2002 (+16.1 in the playoffs).

Stats for the NBA has him at +3.6 and +1.1 (total +4.7) in 2001,+1.4 and +0.6 (+2.1) in 2002. +2.8 and +1.7 (+4.6) in 2000. +2.7 and +2.0 in 99 (+4.7). +3.3 and +2.1 (+5.3) in 1998. +3.7 and +1.6 (+5.3) in 97.

Not really a huge tail-off there until 2002, which jives with his general efficiency and playoff stats.

The Google sites RAPM page has 2002 Malone at -0.7 on offense and +0.2 on defense. They have his NPI RAPM for 2001 at +2.5 and 0.3.

So there's some divergence in the actual totals, but a fairly noticeable trend of his drop-off happening more at 2002 than just after 1998. And that means that he's generally got around 14 years of top-end play... while maintaining starter's minutes. It's a credit to KG that he had the skills to shift himself into more of a defensively-focused player so that he could be very useful to a team that way and with a reduced offensive load in fewer minutes, so there's that angle to consider, but I think Doc's post undersells Malone's offensive utility somewhat.

[Doctor MJ]

2010 Kobe:

+1.3 DRAPM
Lakers on-court defensive rating with Bryant on: 103.0
Lakers on-court defensive rating with Bryant off: 106.4

-3.4

You heard it hear first. Kobe actually peaked as a defender at age 31 in the last year of his prime.

RAPM says so.

Don't expect a coherent response. When someone critiques RAPM, there's always an excuse for why the strange result doesn't count and their point does. It's becoming such a drag on this project.

Sent from my GT-I9300 using RealGM Forums mobile app

People have gone to such great lengths to explain RAPM in this project alone, yet all you see are excuses. To me that's the drag. It feels like you're saying "If you guys had something you'd hit this target" while not pointing to any target, and then bashing the group afterward for not hitting the mark while still not explaining what it is we need to do to hit it.

So, what do I make of Kobe's defensive RAPM in 2010?

Well, the data you're using looks like the scaled PI stuff from my spreadsheet, if that's the case it's influenced by that year first and foremost and then the prior year.

That two year stretch was also the best two-year defensive stretch the Lakers ever had in Kobe's career, and of course also the back-to-back championships that represent Kobe's magnus opus.

So then, is it a coincidence that a player looks best by +/- when his team is at its most effective? Probably not. So yeah, that's what I'd say is going on. This was Kobe in a consistent, very healthy system, and it seems to have had some effect on his efficacy.

And logically, if you're a gambler, sure is nice to have twin tower at your back, no? Might make it so that the team benefits from your good with less consequence from the bad? Is that so weird?

Now though we shouldn't be get carried away. That's less than 1 bucket of impact per game, and he spent virtually his entire career within 1 bucket of neutral. I look at the whole thing and I don't really see the outliers, I just see a guy who was mostly in the roughly neutral range.

[An Unbiased Fan]

You're factually incorrect in how you're describing regression results. This isn't a matter of opinion. Regression, literally, is a technique that takes in a bunch of inputs and results, and identifies which INDIVIDUAL(s) in that group of inputs are most correlated with the given outputs. Again, this isn't something to have a take on. This isn't something to debate. Factually, this is what regression does.

Dude, I'm not arguing the validity of ridge regression. I'm talking about the lineup data samples used as a base in RAPM, which can't be used effectively to derive individual impact. The individual "correlation" is of the lineups the "individual" is in, which is what I already said before. RR is not magic, and simply is incapable of doing what you want it to do. Again, you would need to first get data samples(like synergy) that add individual context. I've been saying this for 4 years since it first emerged on RealGM. But seriously, I have no interests in taking up space on RAPM. The problem with it is simple, which is the use of +/- lineup data only as a base. I only had to challenge the notion that 98/99 KG > 06/07 Kobe earlier, or that 2011 KG had more impact than any version of Kobe. Another gift from RAPM.

(Barkley voice) "Tikhonov is rolling over in his grave, Ernie...".