RealGM 2023 Top 100 Project - #33 (Deadline 5:00AM PST on 10/12/23)

[Owly]

Frazier having the edge in peak, and Stockton the edge in longevity.

I might be inclined to dispute ...
a) the same terminology to descibe each notional advantage
b) the specific phrasing "edge" to describe Stockton's longevity lead.

[WestGOAT]

I'd actually totally buy-in Stockton being a top 20 guy if he'd played consistently at the level he did during the final series of a season like the 1988 and 1989 Playoffs (vs LAL and GSW), but more often than not he did not: the Jazz could have perhaps advanced if his TS% does not plummet (1990 PHO, 1994 HOU, 1996 SEA), or a (lesser) PG counterpart is matching or outplaying him (1991, 1992 POR and 1994,1995 HOU).

What's especially concerning, and based on the superlatives heaped on Stockton, you'd expect him to have outplayed Kenny Smith (where would peak Kenny Smith be ranked?) at least in one of their series. Is that unreasonable to expect from an all-time great?

[MrLurker]

Vote: Stockton, Stockton, Stockton!!!!!!!!!!
Clearly the most underrated player in the project IMO. I’d have him ahead of the player who went in #7 (Wilt) and I’m not even big on longevity!!! All-time assists leader (by a mile). All-time steals leader (by a mile). #5 player in 26 year RAPM looking only at his age 34-40 seasons!!! Don’t just throw that out! He showed greater impact than Shaq or Duncan or a ton of other stars did for their entire careers at an age when Jordan had already retired… twice!!! He’s #3 all-time in VORP. Honestly, I think there’s a very good chance I’m underrating him at #15 on my all-time list.

Alternate: Kawhi
All-time 2 way peak. Will get into it more after Stockton goes which better be this thread!!!

Nominate: Anthony Davis
Top 6 in 3 box score composites between regular season and playoffs with enough defensive impact that he should get the same benefit of the doubt on exceeding his box score numbers as say Patrick Ewing. For all the talk about his injuries, he’s only ever missed one playoff game.

Alternate: Jason Kidd
Only player in the top 20 in VORP other than Stockton who hasn’t been inducted yet. Led 2 teams to the Finals as the CLEAR best player. Finished #2 for MVP. Had a major contribution to a championship team at age 37. Basically did it all between peak and longevity.

I must say I am sympathetic to Unibrodavis's concerns when all-in-one's are used like this - doesn't BPM give guards extra points when they collect stats simply for being guards

[iggymcfrack]

Vote: Stockton, Stockton, Stockton!!!!!!!!!!
Clearly the most underrated player in the project IMO. I’d have him ahead of the player who went in #7 (Wilt) and I’m not even big on longevity!!! All-time assists leader (by a mile). All-time steals leader (by a mile). #5 player in 26 year RAPM looking only at his age 34-40 seasons!!! Don’t just throw that out! He showed greater impact than Shaq or Duncan or a ton of other stars did for their entire careers at an age when Jordan had already retired… twice!!! He’s #3 all-time in VORP. Honestly, I think there’s a very good chance I’m underrating him at #15 on my all-time list.

Alternate: Kawhi
All-time 2 way peak. Will get into it more after Stockton goes which better be this thread!!!

Nominate: Anthony Davis
Top 6 in 3 box score composites between regular season and playoffs with enough defensive impact that he should get the same benefit of the doubt on exceeding his box score numbers as say Patrick Ewing. For all the talk about his injuries, he’s only ever missed one playoff game.

Alternate: Jason Kidd
Only player in the top 20 in VORP other than Stockton who hasn’t been inducted yet. Led 2 teams to the Finals as the CLEAR best player. Finished #2 for MVP. Had a major contribution to a championship team at age 37. Basically did it all between peak and longevity.

I must say I am sympathetic to Unibrodavis's concerns when all-in-one's are used like this - doesn't BPM give guards extra points when they collect stats simply for being guards

It's funny, when Jokic was ranking high in BPM, people were saying the stat was biased in favor of centers. Neither one is true. They estimate roles from their statistical profile and then adjust them. Since center's assists are usually higher value assists than guard assists, they get more credit there. Guards get less credit for defensive rebounds since their defensive rebounds are usually uncontested while their offensive rebounds and blocks are worth more since that's usually something their replacement wouldn't do. Also, point guards get an automatic penalty on defense because they're assumed to be worse defenders on average. It's all to best fit the data. FWIW, here's where the top guards rank in VORP:

3. Stockton
6. CP3
13. Kobe
15. Magic
17. Harden
20. Kidd

Doesn't look very biased for guards to me.

[iggymcfrack]

If BPM was the whole basis of Stockton’s argument, it might be more of an issue. His impact numbers actually greatly outperform his BPM over a 7 season sample. That sample is 3300 minutes more than Embiid’s entire career. It’s very significant! He’s 6th all-time in win shares. The only player that played their whole career in the BPM era who’s ahead behind Stockton in BPM and ahead in win shares is Karl Malone. Like I say about Stockton, the evidence is pretty unanimous. It all points in one direction.

[70sFan]

If BPM was the whole basis of Stockton’s argument, it might be more of an issue. His impact numbers actually greatly outperform his BPM over a 7 season sample. That sample is 3300 minutes more than Embiid’s entire career. It’s very significant! He’s 6th all-time in win shares. The only player that played their whole career in the BPM era who’s ahead behind Stockton in BPM and ahead in win shares is Karl Malone. Like I say about Stockton, the evidence is pretty unanimous. It all points in one direction.

To the direction that he's better than Wilt...?

On a serious note, I think Stockton should be already in. Stockton, Miller and Havlicek are all in similar tier to me, but my criteria puts Kawhi out of top 40.

[70sFan]

Vote: Stockton, Stockton, Stockton!!!!!!!!!!
Clearly the most underrated player in the project IMO. I’d have him ahead of the player who went in #7 (Wilt) and I’m not even big on longevity!!! All-time assists leader (by a mile). All-time steals leader (by a mile). #5 player in 26 year RAPM looking only at his age 34-40 seasons!!! Don’t just throw that out! He showed greater impact than Shaq or Duncan or a ton of other stars did for their entire careers at an age when Jordan had already retired… twice!!! He’s #3 all-time in VORP. Honestly, I think there’s a very good chance I’m underrating him at #15 on my all-time list.

Alternate: Kawhi
All-time 2 way peak. Will get into it more after Stockton goes which better be this thread!!!

Nominate: Anthony Davis
Top 6 in 3 box score composites between regular season and playoffs with enough defensive impact that he should get the same benefit of the doubt on exceeding his box score numbers as say Patrick Ewing. For all the talk about his injuries, he’s only ever missed one playoff game.

Alternate: Jason Kidd
Only player in the top 20 in VORP other than Stockton who hasn’t been inducted yet. Led 2 teams to the Finals as the CLEAR best player. Finished #2 for MVP. Had a major contribution to a championship team at age 37. Basically did it all between peak and longevity.

I must say I am sympathetic to Unibrodavis's concerns when all-in-one's are used like this - doesn't BPM give guards extra points when they collect stats simply for being guards

It's funny, when Jokic was ranking high in BPM, people were saying the stat was biased in favor of centers. Neither one is true. They estimate roles from their statistical profile and then adjust them. Since center's assists are usually higher value assists than guard assists, they get more credit there. Guards get less credit for defensive rebounds since their defensive rebounds are usually uncontested while their offensive rebounds and blocks are worth more since that's usually something their replacement wouldn't do. Also, point guards get an automatic penalty on defense because they're assumed to be worse defenders on average. It's all to best fit the data. FWIW, here's where the top guards rank in VORP:

3. Stockton
6. CP3
13. Kobe
15. Magic
17. Harden
20. Kidd

Doesn't look very biased for guards to me.

I think you missed one little guard from the 1990s at the 2nd spot...

If you think that 7 guards vs 5 centers out of 20 players isn't overrating, then I don't know.

[iggymcfrack]

I must say I am sympathetic to Unibrodavis's concerns when all-in-one's are used like this - doesn't BPM give guards extra points when they collect stats simply for being guards

It's funny, when Jokic was ranking high in BPM, people were saying the stat was biased in favor of centers. Neither one is true. They estimate roles from their statistical profile and then adjust them. Since center's assists are usually higher value assists than guard assists, they get more credit there. Guards get less credit for defensive rebounds since their defensive rebounds are usually uncontested while their offensive rebounds and blocks are worth more since that's usually something their replacement wouldn't do. Also, point guards get an automatic penalty on defense because they're assumed to be worse defenders on average. It's all to best fit the data. FWIW, here's where the top guards rank in VORP:

3. Stockton
6. CP3
13. Kobe
15. Magic
17. Harden
20. Kidd

Doesn't look very biased for guards to me.

I think you missed one little guard from the 1990s at the 2nd spot...

If you think that 7 guards vs 5 centers out of 20 players isn't overrating, then I don't know.

Oops, IDK how I did that, LOL! I certainly noticed MJ there and was thinking about him before I wrote my post out when I was checking how many there would be.

FWIW, we also have 7 guards in the top 20 in this very project: Jordan (3), Magic (10), Steph (11), Kobe (13), West (14), Oscar (15), and CP3 (20). I don’t see 7 as unreasonable. In my own top 20, I’d have Jordan, Curry, CP3, Magic, Stockton, Wade and Oscar, so again 7.

[penbeast0]

I think you missed one little guard from the 1990s at the 2nd spot...

If you think that 7 guards vs 5 centers out of 20 players isn't overrating, then I don't know.

Without commenting on the stat directly, remember that 7 guards is two positions on the court as guys like Jordan, Kobe, and probably Harden are considered SGs.

[trex_8063]

.

I think you missed one little guard from the 1990s at the 2nd spot...

If you think that 7 guards vs 5 centers out of 20 players isn't overrating, then I don't know.

But "guards" is TWO positions (PG and SG). So they're averaging only 3.5 spots per position, vs. 5 for center ("forwards" averaging 4 per position). EDIT: Or 4 centers, and average of 4.5 per forward position, if one counts Duncan as a PF. Either way, the guard positions average the fewest.

EDIT2: I also note that guards (PG or SG) only hold three of the top 12 spots. Aside from Jordan and Stockton [way up in the top 3], the guards mostly occupy the lower ground, with the average place of those seven guards within the top 20 [even inclusive of Jordan and Stockton] is 10.86 (ever so slightly below the halfway point).

[70sFan]

.

I think you missed one little guard from the 1990s at the 2nd spot...

If you think that 7 guards vs 5 centers out of 20 players isn't overrating, then I don't know.

But "guards" is TWO positions (PG and SG). So they're averaging only 3.5 spots per position, vs. 5 for center ("forwards" averaging 4 per position). EDIT: Or 4 centers, and average of 4.5 per forward position, if one counts Duncan as a PF. Either way, the guard positions average the fewest.

EDIT2: I also note that guards (PG or SG) only hold three of the top 12 spots. Aside from Jordan and Stockton [way up in the top 3], the guards mostly occupy the lower ground, with the average place of those seven guards within the top 20 [even inclusive of Jordan and Stockton] is 10.86 (ever so slightly below the halfway point).

I am aware of that, I just combine it with my opinion that bigger player were significantly more impactful on average than smaller guys. If guards have similar distribution to other positions in the top 20, then I think it shows the stat overrates guards.

[trex_8063]

VOTE: John Stockton (again, and again, and again...)

I'll re-iterate [as was outlined in detail in prior threads] the Jazz offense was usually very good to elite in the playoffs during Stockton's prime, and somewhat consistently OUTPERFORMED their rs standard in terms of rORTG. It's been argued that Malone was more essential to this post-season offensive success, and I won't argue or deny that's likely true (though I'll revisit this sentiment below). However, I don't think it would be fair or accurate to characterize Stockton as merely some kind of glorified role player in terms of his importance/contribution to that offensive success.


Regarding his placement in VORP [3rd since 1973]. It's been said this stat favours guards, which I don't believe is the case (I noted in the last thread that the guard positions each have fewer spots in the top 20 all-time than the C position and BOTH forward positions, and that the average guard place within the top 20 is marginally below the average place of the other positions).

It's been said the weighting within BPM is arbitrary and has bias; this has been said about each "all-in-one" metric, actually. No doubt a nugget of truth in that.
However, are ALL of these all-in-ones biased for Stockton's player type? Are ANY of them, truly?

While VORP doesn't cover all of NBA history, I think it would be fair to assume Stockton would STILL rank in the top 10 of all-time even if it did (maybe even top 5-6). He's 8th since 1973 in career BPM, and is the only one in the top 9 not yet voted in (all others were voted inside the top 26).
AND he's 6th all-time in career WS (with a career WS/48 that's 19th all-time [despite a 19-year career]).
And he's top 40 in career PER (a stat that definitely is NOT biased toward his player type), again: despite a 19-year career. If weighted for minutes (sort of like a cumulative GameScore), he's certainly somewhere in the top 25 at least.

Singly, we can point out flaws in each of the various available metrics. But when they ALL seem to be telling us the same/similar thing.....when the summation of all these things appear to be saying this guy should have been off the table a while ago, it begins to feel a bit dubious to argue him out of even top 35 contention.

Particularly considering his impact indicators are essentially just as strong as his box metrics......

RAPM league-rank:
'97 (NPI): 19th (Malone 14th; Hornacek 9th and Ostertag 12th, fwiw; the mathematics perhaps having difficulty parsing out the line-up noise in an NPI sample, and I would speculate Stockton being hurt for babysitting the [bad] 2nd unit)
'98: 7th (Malone 8th)
'99: 11th (Malone 16th; Stockton 7th in NPI [Malone 19th])
'00: 8th (Malone 19th; Stockton 6th in NPI [Malone 45th])
'01 (NPI): 3rd (Malone around 30-35)
'02: 9th (Malone outside the top 100)
'03: 11th (Malone outside top 40)

We have partial (NPI) RAPM for '88 and '91 (admittedly small samples: around 14-17 games), fwiw:
'88: 14th in league (Malone outside top 100)
'91: 8th (Malone outside top 100)

In Ben's AuPM for '94-'96, Malone comes out notably ahead, though Stockton again formidable in his own right: never lower than 16th in the league ['94], and is immediately behind Malone [they are 5th and 6th] in '96.


This is not me trying to disparage Malone, or say that Stockton was the true "star", or the better/more impactful player. This is me merely pointing out that impact signals largely correlate pretty well to his box-based metrics (which again: are friggin' awesome), and that he was likely really friggin' important to his team's success (perhaps not that far behind Malone in importance).


Though this brings me to more broad point that Owly summed up nicely in last thread:

More generally, in my opinion/for me. you're too much weighting what Stockton wasn't, where I'd read his impact profile and production and conclude he was impactful doing the stuff he did.

I'm sure this happens to most players to some degree, but I perceive it's being used more than for most as a means of arguing Stockton out of contention. Stated arguments against him have included.....

*He's not as impactful/important on the Jazz as Karl Malone.

**He's not Steve Nash (in that he didn't "lead" some of greatest playoff offenses of all-time [though again, he did "Robin" TWO >+8 rORTG's and a couple other >+7's in the playoffs], and he couldn't [or at least didn't] call his own number more and put the team on his back for stretches as needed).

***He hardly ever scored 30+ pts.

****He wasn't ever a top-5 player.


Re: He's not as impactful/important on the Jazz as Karl Malone.
Oh....kay??
While I'll again note [fwiw] that there is actually ample supply of data directly contradicting this presumption, I'll nonetheless go along and say it is true (though feel the margin is much smaller than many seem to pretend). However......That he wasn't [quite] as impactful as the guy voted 19th precludes him from being valid at #32 or #33?


Re: He's not Steve Nash.
OK. He couldn't [or didn't] do some of those things listed above.
If we're only going to focus on the things a guy wasn't, I could say Steve Nash couldn't be [or wasn't ever] a positive impact defender.
And further, even if we operate on the presumption that Nash was better, the same basic question as above [with Malone] applies: so he's NOT [quite] the equal of the guy voted in #24; does that mean he's not valid for #32 or 33?


Re: He hardly ever scored 30+ pts.
This one was specifically in contrast to Scottie Pippen. Don't entirely get the point other than it's again picking at something that Stockton is NOT [a volume scorer] as a means of drawing contrast to a more favoured candidate [who, by coincidence, WAS a more voluminous scorer]. But it establishes very little by way of a salient point regarding their offensive value. Shall I point out that Stockton's declined shooting efficiency in the playoffs during his prime (which has been pointed out repeatedly) is still substantially better than that of Scottie Pippen? Shall I point out that his overall creation rates tower above Pippen's? Shall I point out his superior turnover economy?
Or shall I illustrate the arbitrary nature of the argument by, for example, pointing out that Hassan Whiteside or Clint Capela each have multiple years securing more defensive rebounds than any year of Marc Gasol. Does that mean they were better defenders?


Re: He wasn't ever a top-5 player.
This one has been alluded to repeatedly by more than one detractor, most recently [obviously] in comparison to Scottie Pippen:
Scottie Pippen was a top-5 player in the league at least once or twice, Stockton wasn't (and I agree [though I haven't taken a hard look at '89, where he might have a case]).
Though they were arguably top 7-8(ish) in the league an equal number of times, while Stockton was at least a fringe top-10 player probably 2-3 MORE times; and Stockton probably has AT LEAST 50% more seasons as a fringe top-15 player [or better].

But this arbitrary top-5 cut-off seems to say being top-5 is an enormous deal, while being top 7-8 borders on irrelevant. It asserts that the mere difference [in value] of being 4th-5th best in the league (vs 7th-8th) for just two or three seasons, outweighs the difference [in value] between 15th and top 30-60 for a half-dozen years.
I certainly don't agree, at any rate.


At any rate, while he wasn't any of these things, he was nonetheless very impactful via the things that he actually WAS (see above).


Anyway, I'm out of time, but [my goodness] that should suffice for talking points.
Alt vote: Kawhi

Nomination: Elgin Baylor
Alt Nom: Jason Kidd

[Dutchball97]

I think you missed one little guard from the 1990s at the 2nd spot...

If you think that 7 guards vs 5 centers out of 20 players isn't overrating, then I don't know.

But "guards" is TWO positions (PG and SG). So they're averaging only 3.5 spots per position, vs. 5 for center ("forwards" averaging 4 per position). EDIT: Or 4 centers, and average of 4.5 per forward position, if one counts Duncan as a PF. Either way, the guard positions average the fewest.

EDIT2: I also note that guards (PG or SG) only hold three of the top 12 spots. Aside from Jordan and Stockton [way up in the top 3], the guards mostly occupy the lower ground, with the average place of those seven guards within the top 20 [even inclusive of Jordan and Stockton] is 10.86 (ever so slightly below the halfway point).

I am aware of that, I just combine it with my opinion that bigger player were significantly more impactful on average than smaller guys. If guards have similar distribution to other positions in the top 20, then I think it shows the stat overrates guards.

BPM takes position into account (at least they try to) and there are some pretty wild swings you can get by changing nothing except for a players' listed position.

Legitimate concerns about a lack of defensive inputs and them changing the formula because Westbrook broke it aside, wouldn't it make more sense to look at BPM as a positional grade instead of a measure of how much impact a player has hollistically?

It's always something I think about with WOWY-type stats as well. Players where the whole offense or defense is built around (so namely primary playmakers who are serious scoring threats as well or defensive anchors) do really well in these stats but how much of that is their ability as a basketball player and how much is it their role allowing them to have that kind of impact? I think it potentially underrates "plug and play" players who can thrive in any system but have a role that is more easily replaced or isn't as vital to the functioning of a team.

[trex_8063]

I think you missed one little guard from the 1990s at the 2nd spot...

If you think that 7 guards vs 5 centers out of 20 players isn't overrating, then I don't know.

But "guards" is TWO positions (PG and SG). So they're averaging only 3.5 spots per position, vs. 5 for center ("forwards" averaging 4 per position). EDIT: Or 4 centers, and average of 4.5 per forward position, if one counts Duncan as a PF. Either way, the guard positions average the fewest.

EDIT2: I also note that guards (PG or SG) only hold three of the top 12 spots. Aside from Jordan and Stockton [way up in the top 3], the guards mostly occupy the lower ground, with the average place of those seven guards within the top 20 [even inclusive of Jordan and Stockton] is 10.86 (ever so slightly below the halfway point).

I am aware of that, I just combine it with my opinion that bigger player were significantly more impactful on average than smaller guys. If guards have similar distribution to other positions in the top 20, then I think it shows the stat overrates guards.

I guess it depends on what you call "similar distribution". As per above, the guards are under-represented relative to the "bigger" players (ALL positions of "bigger" players), though they are close.

Given the more modern era [which comprises a large chunk of the VORP sample] has trended toward perimeter-based play, I wouldn't call that "overrating" guards, personally.

[tsherkin]

As per above, the guards are under-represented relative to the "bigger" players (ALL positions of "bigger" players), though they are close.

Are they? Or is it reflective of a historical context where it has been harder for guards to exert the same level of impact as larger players?

[Ambrose]

This is a tough one. Stockton, Miller and Kawhi are all guys I'd consider here. Stockton and Kawhi are two opposite sides of the spectrum, and Miller is the middle ground choice.

Considering that Miller was the leader of a handful of very good teams, he was a guy who blended in easily when his casts improved, he had excellent durability and longevity, and is one of the greatest postseason risers ever...I'll side with him.

Vote: Reggie Miller
Alternate vote: John Stockton
Nominate: Jason Kidd

[trex_8063]

As per above, the guards are under-represented relative to the "bigger" players (ALL positions of "bigger" players), though they are close.

Are they? Or is it reflective of a historical context where it has been harder for guards to exert the same level of impact as larger players?

I don't want to belabour the point and/or derail the thread on this, but I was speaking in an absolute, numerical sense: they are not represented as heavily [numerically] as the other positions.

Yes, I believe it has been historically more difficult for guards to exert the same level of impact and/or statistical imprint as larger players. And from the standpoint of acknowledging that, the VORP top-20 standings could be said or argued to accurately reflect that legacy (i.e. we expect [based on "historical context"] that they will be less represented on this listing than the other positions.......and that is, in fact, what we see).

I guess it's only a question of if you feel they are "under-represented" [again: in a strictly numerical sense] enough to accurately reflect historical context.

I don't feel---based on their representation, which lags behind other positions---it can be said with certainly that it overrates guards, as has been said. Particularly given a LARGE proportion of the player-seasons which have occurred since 1973 occurred in an era that trended toward perimeter-based play/dominance (and diminished the value of the traditional low-post big).

[tsherkin]

I don't want to belabour the point and/or derail the thread on this, but I was speaking in an absolute, numerical sense: they are not represented as heavily [numerically] as the other positions.

Yes, I believe it has been historically more difficult for guards to exert the same level of impact and/or statistical imprint as larger players. And from the standpoint of acknowledging that, the VORP top-20 standings could be said or argued to accurately reflect that legacy (i.e. we expect [based on "historical context"] that they will be less represented on this listing than the other positions.......and that is, in fact, what we see).

I guess it's only a question of if you feel they are "under-represented" [again: in a strictly numerical sense] enough to accurately reflect historical context.

I don't feel---based on their representation, which lags behind other positions---it can be said with certainly that it overrates guards, as has been said. Particularly given a LARGE proportion of the player-seasons which have occurred since 1973 occurred in an era that trended toward perimeter-based play/dominance (and diminished the value of the traditional low-post big).

Right on, appreciate the clarification.

[Dutchball97]

Vote: Kawhi Leonard - My post is probably going to be a bit boring as neither my vote nor my alternate got in last round. That said I see Kawhi as having the highest peak here by a solid margin, outside of Frazier I don't think the other nominees come particularly close. Kawhi was in my opinion the best player in the world in 2019 and he has enough seasons of being an elite player in the regular season before taking it up to all-time levels in the play-offs that I am confident we're not looking at a fluke player who had only 1 or 2 high level seasons before fizzling out. While Kawhi might not have the best longevity, he has still played more minutes than Jokic, who is already in. This is especially prominent with Kawhi roughly playing double the minutes and games as Jokic in the post-season. On the stage where it matters most, Kawhi has no longevity issues at all as evidenced by him being 54th in career play-off minutes (ahead of guys like Oscar, Moses, Nash and Barkley) and 17th in career play-off Win Shares. It's also telling that longevity giant, Karl Malone, is 16th on that list with just 0.12 more career play-off WS than Kawhi despite playing 56 more games(193 and 137 respectively).

Alternate Vote: Walt Frazier - Kind of a similar case to Kawhi and with Frazier having over 2 POY shares (including 4 straight top 3 placements from 1970-1973) I don't think there are many doubts to just how good Frazier was in the early 70s. Like Kawhi, he's also a very strong play-off performer. Out of the 8 post-seasons (1968-1975) Frazier played in, he had good to great runs in all but his rookie season.

For runoff purposes I'd rank the rest of the nominees as John Stockton > John Havlicek > Reggie Miller.

Nominate: Anthony Davis - Can't say I'm entirely sure about this but despite there being plenty of candidates around, none of them really jump out at me as guys that just have to make the list asap. AD has suffered a lot of injuries but that didn't stop him from racking up quite a career already. I do think him playing only a handful of post-seasons hurts him compared to most of his contemporaries but it's easy to forget just how well Davis has performed every time he did get a chance to show off on the biggest stage.

[DSMok1]

Going to clear up a few questions about BPM/VORP here:

I must say I am sympathetic to Unibrodavis's concerns when all-in-one's are used like this - doesn't BPM give guards extra points when they collect stats simply for being guards

It's funny, when Jokic was ranking high in BPM, people were saying the stat was biased in favor of centers. Neither one is true. They estimate roles from their statistical profile and then adjust them. Since center's assists are usually higher value assists than guard assists, they get more credit there. Guards get less credit for defensive rebounds since their defensive rebounds are usually uncontested while their offensive rebounds and blocks are worth more since that's usually something their replacement wouldn't do. Also, point guards get an automatic penalty on defense because they're assumed to be worse defenders on average. It's all to best fit the data. FWIW, here's where the top guards rank in VORP:

3. Stockton
6. CP3
13. Kobe
15. Magic
17. Harden
20. Kidd

Doesn't look very biased for guards to me.

I would say BPM is fairly balanced guards vs. big men. There are different weights for many of the box score statistics based on the estimated position of the player and the estimated offensive role of the player.

BPM IS biased towards players that produce good box scores. The argument should be primarily about whether the player's box score statistics accurately reflect the player's actual impact.

As a rule of thumb, I would say the box score captures 80% of a player's offensive impact and 50% of a player's defensive impact. That is based on numerous regressions I have run onto RAPM-style metrics.

I personally would argue that John Stockton had a profile that would be overrated by the box score. His steals and his assists both tended to overstate his impact on the game in those areas, I feel. And his BPM is reflective of that. In other words--some of the credit that his assists and steals are collecting for his BPM/VORP should actually have gone to his teammates.

But "guards" is TWO positions (PG and SG). So they're averaging only 3.5 spots per position, vs. 5 for center ("forwards" averaging 4 per position). EDIT: Or 4 centers, and average of 4.5 per forward position, if one counts Duncan as a PF. Either way, the guard positions average the fewest.

EDIT2: I also note that guards (PG or SG) only hold three of the top 12 spots. Aside from Jordan and Stockton [way up in the top 3], the guards mostly occupy the lower ground, with the average place of those seven guards within the top 20 [even inclusive of Jordan and Stockton] is 10.86 (ever so slightly below the halfway point).

I am aware of that, I just combine it with my opinion that bigger player were significantly more impactful on average than smaller guys. If guards have similar distribution to other positions in the top 20, then I think it shows the stat overrates guards.

BPM takes position into account (at least they try to) and there are some pretty wild swings you can get by changing nothing except for a players' listed position.

Legitimate concerns about a lack of defensive inputs and them changing the formula because Westbrook broke it aside, wouldn't it make more sense to look at BPM as a positional grade instead of a measure of how much impact a player has hollistically?

It's always something I think about with WOWY-type stats as well. Players where the whole offense or defense is built around (so namely primary playmakers who are serious scoring threats as well or defensive anchors) do really well in these stats but how much of that is their ability as a basketball player and how much is it their role allowing them to have that kind of impact? I think it potentially underrates "plug and play" players who can thrive in any system but have a role that is more easily replaced or isn't as vital to the functioning of a team.

A couple of points on this: you cannot change a player's BPM by changing their listed position. Their position is estimated based purely on box score inputs; this was part of the development of the BPM calculation itself. Their offensive role is separately estimated based on box score inputs as well.

There were significant issues with the first version of BPM; there was some overfitting going on that made the regression not robust. That is significantly better now, but it is not perfect. There are a couple of additional tweaks that need to be made, but I haven't gotten to it yet.

____
Just think of BPM as a reasonable approximation of what the box score + the overall team quality says about a player. Remember that is blind to around 20% of offense and 50% of defense.

Remember also that VORP is using a replacement level of -2, which is probably not appropriate for a top 100 project. When looking at championship impact, a replacement level of 0 or even +1 would be better. This is also a big swing in Stockton's favor, as he had a lower peak but great longevity.

That said, I think I would still take Stockton over the others nominated....