RealGM

Single season peak only which ones peaked higher?

1. Mitch Richmond
2. Chris Mullin
3. Eddie Jones
4. Joe Dumars
5. Peja Stojakovic
6. Shawn Marion
7. Detlef Schrempf
8. Andrei Kirilenko

I'd have AK47 as the highest peak, followed closely by Marion.

Colbinii wrote:I'd have AK47 as the highest peak, followed closely by Marion.

AK47
Marion
Middleton
—
then some order of the rest of the bunch?

Honestly, I'd say all of them.

Richmond did his best on crap teams, but there was one year where he floated a terrible supporting cast to 34 wins (19th of 29 offensively) while being 4th in the league in ppg on good all-around efficiency, and by memory/reputation was passable/decent(ish) on defense (not sure I'd say Middleton is any better on D).

Eddie Jones was an outstanding defensive wing whose impact signals are consistently ABOVE his box-based aggragates.......though even his box-based aggragates are marginally better than Middleton's.

Detlef was a big guy with an extremely diverse skillset: he could stretch the floor (as a combo forward), score efficiently, distribute the ball, rebound extremely well (for a guy playing some SF), and play passable defense otherwise, too.

Joe Dumars has been historically overrated [imo], and particularly his defense; but still we're talking about a guy with comparable box-based metrics with defense that is notably above what the box suggests [note for example he's a career -0.7 DBPM and 110 DRtg......he was certainly substantially better than what that indicates, by way of being good at the things that do not show up in the box]. He might be the one that's closest to Middleton's peak, perhaps near a tie; though gun to my head I'd likely give the nod to Dumars.


I frankly think don't think Middleton was even all that close to any of the others.....

Peja at his peak was the leading scorer [2nd in the entire league] on frankly ridiculous +10.8% rTS, as co-anchor of the 2nd-best offense in the league.

Marion was never a big on-ball scorer, but he managed to score pretty well by being an elite cutter or baseline runner, elite transition finisher, and decent spot-up shooter (and good FT shooter) despite his weird looking shot. Additionally he was an excellent and SUPER-versatile defender (that he never received an All-Defensive nod is arguable the biggest career award snub in history), and one of the very best rebounding SF's in NBA history, who also played very well within his niche/skillset (and thus turned it over very rarely). These things add up.

And as far as Mullin and Kirilenko go........I think one must have an over-inflated opinion of Middleton, and/or an inappropriately low opinion of Mullin and AK47 to even ask. Because Middleton was never close to their peaks.

I don't see Middleton being able to challenge Dumars, AK47, Marion and Peja. All of those guys had top tier impact to them. Peja has a nice + to the O-Rating game, Marion is a Top 100 type, Dumars is a top defender in the stacked 80's time, and AK-47 had all round numbers for his day.

The best Middleton can do is land #5, but the good thing about Middleton if you watch him closely is he has a nice shot and regularly delivers as the #2. That being said he deserves some criticism if Milwaukee only nets 1 title with Giannis. He had 4 nice games vs the Pacers this yr, #2's like that aren't easy to find.

Looking at this in an absolute sense, what I like about Middleton is that he maintains the best in the postseason of… basically all these players, save for maybe Chris Mullin. Not saying that makes him the best on its own, and a lot of this depends on team structure too — Marion and Kirilenko would both be worse as dedicated small forwards next to Giannis, although peak Kirilenko probably maintains enough dynamism to be an improvement over Middleton regardless — but I think he is being a bit too easily dismissed here. Especially when I would take him as a defender over Peja, Mullin, Richmond, Detlef, and maybe even Dumars. Again, talking in the absolute; relative to his league, all these names save for Detlef had better top fifteen cases in their time.

Now, as a team’s best player, or even just as a first option, none of these names are taking you too far; however, in those circumstances I would only take Kirilenko, Mullin, and Richmond if I wanted to maybe manage one series win. And players like Marion and Jones obviously gain more comparative value the less you need to rely on their scoring, although I think Jones at least properly needs to be a third scoring option to get to the point where I would prefer him over Middleton in the postseason.

I am very low on postseason Peja here, in apparently stark contrast with some of the people here treating him as one of the easier selections ahead. Exploitable defender who could be relatively limited as a scorer when teams dedicated themselves to guarding him close, and that “gravity” still requires strong teammates to properly take advantage. He is also the most likely to have his value capped by periodic cold streaks where he has too little else to offer. Everyone else in the this group is either a competent playmaker or an elite defender (Marion). I would much rather have someone like Klay Thompson, and I think people would feel pretty split on his case against Middleton if he had been included here.

trex_8063 wrote:Honestly, I'd say all of them.

Richmond did his best on crap teams, but there was one year where he floated a terrible supporting cast to 34 wins (19th of 29 offensively) while being 4th in the league in ppg on good all-around efficiency, and by memory/reputation was passable/decent(ish) on defense (not sure I'd say Middleton is any better on D).

Eddie Jones was an outstanding defensive wing whose impact signals are consistently ABOVE his box-based aggragates.......though even his box-based aggragates are marginally better than Middleton's.

Detlef was a big guy with an extremely diverse skillset: he could stretch the floor (as a combo forward), score efficiently, distribute the ball, rebound extremely well (for a guy playing some SF), and play passable defense otherwise, too.

Joe Dumars has been historically overrated [imo], and particularly his defense; but still we're talking about a guy with comparable box-based metrics with defense that is notably above what the box suggests [note for example he's a career -0.7 DBPM and 110 DRtg......he was certainly substantially better than what that indicates, by way of being good at the things that do not show up in the box]. He might be the one that's closest to Middleton's peak, perhaps near a tie; though gun to my head I'd likely give the nod to Dumars.


I frankly think don't think Middleton was even all that close to any of the others.....

Peja at his peak was the leading scorer [2nd in the entire league] on frankly ridiculous +10.8% rTS, as co-anchor of the 2nd-best offense in the league.

Marion was never a big on-ball scorer, but he managed to score pretty well by being an elite cutter or baseline runner, elite transition finisher, and decent spot-up shooter (and good FT shooter) despite his weird looking shot. Additionally he was an excellent and SUPER-versatile defender (that he never received an All-Defensive nod is arguable the biggest career award snub in history), and one of the very best rebounding SF's in NBA history, who also played very well within his niche/skillset (and thus turned it over very rarely). These things add up.

And as far as Mullin and Kirilenko go........I think one must have an over-inflated opinion of Middleton, and/or an inappropriately low opinion of Mullin and AK47 to even ask. Because Middleton was never close to their peaks.

There's part of me that agrees.

The other part says look at Middleton's long-term impact signal via on-off or long term RAPM numbers. Look at his box peak. I have also been under the impression (think via Ben Taylor) that his boxscore production showed a capacity to jump with GA off the floor.

On the box-side for RS peak I would suggest Middleton on aggregate looks marginally better than Jones, by the Reference metrics, depending how one weights them. Peak Peja probably has an advantage in this regard but ... even allowing for some wiggle room on peak Stojakovic's defense is probably reckoned to be behind Middleton's.

In terms of full RS numbers that Middleton peak is outlier.

I don't have a hard and fast position, haven't really thought about it and don't go in for peaks that much ... these are good players so there'd be no shame in being behind them ... I just didn't see much acknowledgement of Middleton and wonder if he's somewhat "unsung".

I'd rank them more like
Mullin/AK47
Marion/Dumars
Peja/Mitch/Middleton
Schrempf/Eddie

I think Khris has to get moved up just due to how good he was in that title run. Otherwise I'd have him in the last tier.

Colbinii wrote:I'd have AK47 as the highest peak, followed closely by Marion.

Not as a wing

I think Middleton is better than Schrempf, maybe Jones.

Owly wrote:There's part of me that agrees.

The other part says look at Middleton's long-term impact signal via on-off or long term RAPM numbers. Look at his box peak. I have also been under the impression (think via Ben Taylor) that his boxscore production showed a capacity to jump with GA off the floor.

On the box-side for RS peak I would suggest Middleton on aggregate looks marginally better than Jones, by the Reference metrics, depending how one weights them. Peak Peja probably has an advantage in this regard but ... even allowing for some wiggle room on peak Stojakovic's defense is probably reckoned to be behind Middleton's.

In terms of full RS numbers that Middleton peak is outlier.

I don't have a hard and fast position, haven't really thought about it and don't go in for peaks that much ... these are good players so there'd be no shame in being behind them ... I just didn't see much acknowledgement of Middleton and wonder if he's somewhat "unsung".

Re: the bolded statements....

I think you overlook the limited minutes Middleton was playing on this outlier statistical peak. He was <30 mpg that year.....which means he's basically never playing fatigued, or pacing himself so he doesn't become fatigued, and rarely holding back the aggression due to foul trouble (because he rarely plays enough minutes to get into foul trouble in the first place), etc.

Compared that to Eddie Jones' statistical peak, where he's playing like 39 mpg (even accounting for era differences, it's the [scaled] difference of like a half of a quarter), and still box-aggragates that are near to what Middleton did in the rs [in just 29.9 mpg]:

Khris - 21.1 PER, .205 WS/48, +4.1 BPM
Eddie - 19.9 PER, .172 WS/48, +4.9 BPM


Further, if you look at the playoff performance of Middleton in said peak statistical year, he falls off a cliff from his rs standard: his TS% dropping by -11.2%, for example. Although he's finally playing bigger minutes (35.5 mpg in the playoffs), his metrics drop to 13.5 PER, .043 WS/48, and +/- 0.0 BPM.
Meanwhile, Jones does not see much fall-off [in smaller sample], with 18.4 PER, .152 WS/48, and +4.6 BPM (and that while playing nearly 43 mpg).

I'll additionally note Eddie's PI RAPM [playoff included] in his peak year was +6.33 (5th in the league). Fairly sure Middleton did not approach that in '20 (or perhaps in any year).


Any other year (aside from '20) we look at for Middleton is considerably less impressive statistically in the rs [as you noted: this was an "outlier"].

So respectfully, I still think Jones edges him for peak.

trex_8063 wrote:

Owly wrote:There's part of me that agrees.

The other part says look at Middleton's long-term impact signal via on-off or long term RAPM numbers. Look at his box peak. I have also been under the impression (think via Ben Taylor) that his boxscore production showed a capacity to jump with GA off the floor.

On the box-side for RS peak I would suggest Middleton on aggregate looks marginally better than Jones, by the Reference metrics, depending how one weights them. Peak Peja probably has an advantage in this regard but ... even allowing for some wiggle room on peak Stojakovic's defense is probably reckoned to be behind Middleton's.

In terms of full RS numbers that Middleton peak is outlier.

I don't have a hard and fast position, haven't really thought about it and don't go in for peaks that much ... these are good players so there'd be no shame in being behind them ... I just didn't see much acknowledgement of Middleton and wonder if he's somewhat "unsung".

Re: the bolded statements....

I think you overlook the limited minutes Middleton was playing on this outlier statistical peak. He was <30 mpg that year.....which means he's basically never playing fatigued, or pacing himself so he doesn't become fatigued, and rarely holding back the aggression due to foul trouble (because he rarely plays enough minutes to get into foul trouble in the first place), etc.

Compared that to Eddie Jones' statistical peak, where he's playing like 39 mpg (even accounting for era differences, it's the [scaled] difference of like a half of a quarter), and still box-aggragates that are near to what Middleton did in the rs [in just 29.9 mpg]:

Khris - 21.1 PER, .205 WS/48, +4.1 BPM
Eddie - 19.9 PER, .172 WS/48, +4.9 BPM


Further, if you look at the playoff performance of Middleton in said peak statistical year, he falls off a cliff from his rs standard: his TS% dropping by -11.2%, for example. Although he's finally playing bigger minutes (35.5 mpg in the playoffs), his metrics drop to 13.5 PER, .043 WS/48, and +/- 0.0 BPM.
Meanwhile, Jones does not see much fall-off [in smaller sample], with 18.4 PER, .152 WS/48, and +4.6 BPM (and that while playing nearly 43 mpg).

I'll additionally note Eddie's PI RAPM [playoff included] in his peak year was +6.33 (5th in the league). Fairly sure Middleton did not approach that in '20 (or perhaps in any year).


Any other year (aside from '20) we look at for Middleton is considerably less impressive statistically in the rs [as you noted: this was an "outlier"].

So respectfully, I still think Jones edges him for peak.

Your points aren’t wrong … and I don’t have a strong opinion on rank … but you said

trex_8063 wrote:though even his box-based aggragates are marginally better than Middleton's.

So my primary focus here is on the box side, for peak.

I, personally, would tend to take that as rate in the first instance because minutes aren’t under a player’s control and that then opens another rabbit hole about the correct “replacement” level or whatever level fills alternate minutes.

Secondarily Milwaukee are in a second straight year above 8 SRS (but not defending champs), they’re blowing teams out. I believe that, and I’m open to being wrong, they aren’t pacing him as he can’t do it himself and will only go all out or fall off a cliff but rather he’s very close to their mpg leader; they’re racking up quite a lot of 20+ point wins … they don’t need their top guys out there 40 minutes because they are blowing teams out.

I would tend to look either RS or playoffs (mainly RS because of the bigger sample or else maybe one would take Tracy Murray’s peak over all these guys) … I’m not typically that bothered about them lining up … I suspect single year playoff stuff is heavily influenced by chance, competition etc. I may be in a minority here. I believe you’ve operated on a similar principle for peaks with different metrics for different years in the past. It depends what one wants to measure, there’s a place for both. But Middleton’s playoff production peak is higher too (I would suggest, depending on metric/weighting of choice). So whilst I’ve seen Jones take what I consider unfair flack for his playoffs … if his rate box-peak is better because of postseason it’s because he happened to line up his strong years better … to an extent that’s more likely with him having more strong production years but now we’re arguably moving away from peak to prime. For non-box side this probably matters more where you can say ... this impact isn't a fluke, it's signal, not noise. But the production is there (and, at the margins, as noted, I believe it has been suggested KM's production was deflated with GA, and up in the time he was off).


PI RAPM … perfectly good tool as to overall. It’s (as you will know - assume this is in support of your general case, not the box-side point) not a box-measure. Even just for the year versus year comp I don’t love it. PI is generally more accurate. But the priors mean you’re not comparing the years on an even ground so if possible I’d seek NPI.

I wouldn’t claim to know the best source (other than not gitlab) but fwiw the first source I came across foe 2020 was NBAShotcharts.com 4 factors RAPM spreadsheets and fwiw the first column variant is Luck Adjusted RAPM … KM is 3rd (behind GA and Kawhi) … there is a column later on just called RAPM where he ranks 17th.

JE’s 97-19 spreadsheet has 2000 Jones as 5th as do A Screaming’s RPI and NPI.

I’m not in a position to compare sources merits as I said. I guess it’s nice that Jones’s rank is secure across the variants/sources … to me, at this time, your sureness on KM not ranking in such a position seems misplaced.

Owly wrote:

trex_8063 wrote:though even his box-based aggragates are marginally better than Middleton's.

So my primary focus here is on the box side, for peak.

I, personally, would tend to take that as rate in the first instance because minutes aren’t under a player’s control and that then opens another rabbit hole about the correct “replacement” level or whatever level fills alternate minutes.

Fair enough, and I guess I could have been more explicitly clear of what I meant by "box-based aggragates". As I've said multiple times in multiple other places in the past: I think it's very important to scrutinize playing time when comparing by way of any rate metrics.
Hyperbolic example would be if we were comparing two guys who had similar(ish) [and fairly impressive] metrics, but one guy is playing like 11 mpg, while the other is playing 33 mpg (this might be like late-career JaVale McGee vs some All-Star(ish) player). Even though their metrics may be similar [maybe even McGee's look marginally better, as he often looks very good by the box], I'd be pretty comfortable declaring---with no additional information---that the 33 mpg guy is doing a helluva lot more for his team, and/or is the better, more valuable player.

Similar principle would generally apply [just to a lesser degree] in a difference of 9 mpg (or a little less in era-scaled terms).

Owly wrote:Secondarily Milwaukee are in a second straight year above 8 SRS (but not defending champs), they’re blowing teams out. I believe that, and I’m open to being wrong, they aren’t pacing him as he can’t do it himself and will only go all out or fall off a cliff but rather he’s very close to their mpg leader; they’re racking up quite a lot of 20+ point wins … they don’t need their top guys out there 40 minutes because they are blowing teams out.

And this may often be the case. otoh, it's worth mentioning that the fact of being on a +8 SRS team that is frequently blowing teams out is also likely inflating [relative to '00 Jones, who played for a 49-win, +2.33 SRS team] Middleton's WS/48 and BPM, as those things are influenced/curved by the team success [pt differential].
If two players have the exact same rate of production in ALL box components, with the exact same shooting efficiency......but one does it for a +2 SRS team while the other for a +8 SRS team, the latter is likely to be awarded a higher WS/48 and BPM.

Owly wrote:I would tend to look either RS or playoffs (mainly RS because of the bigger sample or else maybe one would take Tracy Murray’s peak over all these guys)

idk, I'm more of a mind to consider both, particularly if the rs looks close (as it does in the case of Jones/Middleton). And sample size would of course always be a consideration, which is why we'd not make massive adjustments on Tracy Murray (since we're talking about a 3-game sample [where he played just 29 mpg, fwiw], and which appears entirely out of character [unsustainable] for him).

Owly wrote:PI RAPM … perfectly good tool as to overall. It’s (as you will know - assume this is in support of your general case, not the box-side point) not a box-measure......

You assume correctly.

Owly wrote:Even just for the year versus year comp I don’t love it. PI is generally more accurate. But the priors mean you’re not comparing the years on an even ground so if possible I’d seek NPI.

I'd see that as more of an issue if the preceeding years were prime years for one player, but NOT for the other. This is not the case for the comparison in question, however.

Owly wrote:I wouldn’t claim to know the best source (other than not gitlab) but fwiw the first source I came across foe 2020 was NBAShotcharts.com 4 factors RAPM spreadsheets and fwiw the first column variant is Luck Adjusted RAPM … KM is 3rd (behind GA and Kawhi) … there is a column later on just called RAPM where he ranks 17th.

JE’s 97-19 spreadsheet has 2000 Jones as 5th as do A Screaming’s RPI and NPI.

I’m not in a position to compare sources merits as I said. I guess it’s nice that Jones’s rank is secure across the variants/sources … to me, at this time, your sureness on KM not ranking in such a position seems misplaced.

Fair point; I wasn't aware of Middleton's strong standing in NPI that year.
That said, Eddie Jones is still right there with him in NPI.......and I'd again note the discrepancy in playing time (given RAPM is also a rate metric): 29.9 mpg in rs and 35.5 in ps for Middleton; 39.0 mpg in rs and 42.8 in ps for Jones. Looked at in pts added terms, the edge would go to Eddie Jones, even if going by a scaled NPI.


idk.....perhaps I'm selling Middleton short; I still feel like I'd go with Jones by a hair, though it's sort of situational. If my biggest need was a secondary scorer, and a guy who was capable of occasionally calling his own number and making a play from time to time, I'd want Middleton. If my needs leaned more toward perimeter defense combined with a good tertiary scorer, I'd go with Jones.
And on the whole [in a vacuum], or if asked who I think brought more value overall to their respective teams in those years, I think it's Jones slightly.

And you may have the last word, if you like.

Dr Positivity wrote:

Colbinii wrote:I'd have AK47 as the highest peak, followed closely by Marion.

Not as a wing

I think Middleton is better than Schrempf, maybe Jones.

As a player they did, is that not the question of the thread?

Or are we asking "Who would be best as a shot creator next to Giannis, Jrue and Brook?"

Colbinii wrote:

Dr Positivity wrote:

Colbinii wrote:I'd have AK47 as the highest peak, followed closely by Marion.

Not as a wing

I think Middleton is better than Schrempf, maybe Jones.

As a player they did, is that not the question of the thread?

Or are we asking "Who would be best as a shot creator next to Giannis, Jrue and Brook?"

Once it was listed in the OP I don't have a problem with people ranking him, though spiritually it feels a bit wrong to me as I'd have the SF version of AK during the Deron-Boozer-Okur years last on this list easily, maybe at the time we were wrong and ppg gazing to to consider the Jazz big 3 to have Okur as the 3rd guy and not him (although the value of Okur's spacing would've been underappreciated at the time), but I still think AK is more like their Lopez than Middleton at that point being generous. I'm surprised he still had over 5 BPM in 08 because it didn't feel that way at the time.

But I take it back because now that I look at it, 06 AK was playing SF beside Boozer and Okur/Collins, and was dominant like the previous years. So his random decline from 06 to 07 may not have been a fit thing unless for some reason the rise of Deron affected him, or Sloan getting to go all in on PG/PFs PNRs again. I started watching basketball seriously in 06-07 having been too much of a casual to see the post Stockton/Malone Jazz and the mythical athletic freak version of AK seemed gone to me, I always found it was strange that someone could decline that much athletically at like 25/26.

trex_8063 wrote:

Owly wrote:

trex_8063 wrote:though even his box-based aggragates are marginally better than Middleton's.

So my primary focus here is on the box side, for peak.

I, personally, would tend to take that as rate in the first instance because minutes aren’t under a player’s control and that then opens another rabbit hole about the correct “replacement” level or whatever level fills alternate minutes.

Fair enough, and I guess I could have been more explicitly clear of what I meant by "box-based aggragates". As I've said multiple times in multiple other places in the past: I think it's very important to scrutinize playing time when comparing by way of any rate metrics.
Hyperbolic example would be if we were comparing two guys who had similar(ish) [and fairly impressive] metrics, but one guy is playing like 11 mpg, while the other is playing 33 mpg (this might be like late-career JaVale McGee vs some All-Star(ish) player). Even though their metrics may be similar [maybe even McGee's look marginally better, as he often looks very good by the box], I'd be pretty comfortable declaring---with no additional information---that the 33 mpg guy is doing a helluva lot more for his team, and/or is the better, more valuable player.

Similar principle would generally apply [just to a lesser degree] in a difference of 9 mpg (or a little less in era-scaled terms).

Owly wrote:Secondarily Milwaukee are in a second straight year above 8 SRS (but not defending champs), they’re blowing teams out. I believe that, and I’m open to being wrong, they aren’t pacing him as he can’t do it himself and will only go all out or fall off a cliff but rather he’s very close to their mpg leader; they’re racking up quite a lot of 20+ point wins … they don’t need their top guys out there 40 minutes because they are blowing teams out.

And this may often be the case. otoh, it's worth mentioning that the fact of being on a +8 SRS team that is frequently blowing teams out is also likely inflating [relative to '00 Jones, who played for a 49-win, +2.33 SRS team] Middleton's WS/48 and BPM, as those things are influenced/curved by the team success [pt differential].
If two players have the exact same rate of production in ALL box components, with the exact same shooting efficiency......but one does it for a +2 SRS team while the other for a +8 SRS team, the latter is likely to be awarded a higher WS/48 and BPM.

Owly wrote:I would tend to look either RS or playoffs (mainly RS because of the bigger sample or else maybe one would take Tracy Murray’s peak over all these guys)

idk, I'm more of a mind to consider both, particularly if the rs looks close (as it does in the case of Jones/Middleton). And sample size would of course always be a consideration, which is why we'd not make massive adjustments on Tracy Murray (since we're talking about a 3-game sample [where he played just 29 mpg, fwiw], and which appears entirely out of character [unsustainable] for him).

Owly wrote:PI RAPM … perfectly good tool as to overall. It’s (as you will know - assume this is in support of your general case, not the box-side point) not a box-measure......

You assume correctly.

Owly wrote:Even just for the year versus year comp I don’t love it. PI is generally more accurate. But the priors mean you’re not comparing the years on an even ground so if possible I’d seek NPI.

I'd see that as more of an issue if the preceeding years were prime years for one player, but NOT for the other. This is not the case for the comparison in question, however.

Owly wrote:I wouldn’t claim to know the best source (other than not gitlab) but fwiw the first source I came across foe 2020 was NBAShotcharts.com 4 factors RAPM spreadsheets and fwiw the first column variant is Luck Adjusted RAPM … KM is 3rd (behind GA and Kawhi) … there is a column later on just called RAPM where he ranks 17th.

JE’s 97-19 spreadsheet has 2000 Jones as 5th as do A Screaming’s RPI and NPI.

I’m not in a position to compare sources merits as I said. I guess it’s nice that Jones’s rank is secure across the variants/sources … to me, at this time, your sureness on KM not ranking in such a position seems misplaced.

Fair point; I wasn't aware of Middleton's strong standing in NPI that year.
That said, Eddie Jones is still right there with him in NPI.......and I'd again note the discrepancy in playing time (given RAPM is also a rate metric): 29.9 mpg in rs and 35.5 in ps for Middleton; 39.0 mpg in rs and 42.8 in ps for Jones. Looked at in pts added terms, the edge would go to Eddie Jones, even if going by a scaled NPI.


idk.....perhaps I'm selling Middleton short; I still feel like I'd go with Jones by a hair, though it's sort of situational. If my biggest need was a secondary scorer, and a guy who was capable of occasionally calling his own number and making a play from time to time, I'd want Middleton. If my needs leaned more toward perimeter defense combined with a good tertiary scorer, I'd go with Jones.
And on the whole [in a vacuum], or if asked who I think brought more value overall to their respective teams in those years, I think it's Jones slightly.

And you may have the last word, if you like.

I'll try to be brief. Much to agree to but I'll clarify and something like argue in one place.

The debate would on point 3. Between the response and the chopping of the post it might look as though I don't look at playoffs (or else not at RS). The point was I'd tend to separate the two for the reasons outlined after the cut. And as I said I think you've done something like the same principle with splitting peaks across years by metric (think it was you, and someone didn't like it for a peaks project). I think it's all covered in the post. Jones has better alignment, more good years. If alignment matters that's fine, sometimes it does. Middleton probably has the better individual box peaks for RS and for playoffs.

Murray, fwiw is just throwing a bit of light ... or a grenade ... into the what to do about playoffs issue. Some people have no sample, some will have small ones (actually everyone will, but some smaller than others), uneven competition (working against Murray fwiw) ... I don't know if there's a good, systematic approach. You speak of adjusting which suggests RS is the baseline ... is it something systematic? As I say, it's messy. You also allude to sustainability and ... when one becomes convinced on that and how one adjusts is a problem. Fwiw, in the playoff Middleton '20 case those don't look like numbers that sustain (though that is sort of just saying they’re bad).

As far as box-metrics: Agree minutes matter to value. Player goodness … maybe? Depends why (see Bucks context). If it’s about value contributed then McGee … might illustrate your point regarding minutes but as far as you’re regarding goodness, as you say he’s a less than his production player in a way I don’t think KM is. Maybe one could argue Jones as particularly strong though I think KM looks good in long-term RAPMs.

in a difference of 9 mpg (or a little less in era-scaled terms)

I would tend to lead (or go entirely) with the latter. There’s little good reason to push guys for circa 40mpg in the RS and teams realize this now.

Being on a good team does help WS/48, BPM. I’d just say you defined terms knowing it’s baked in, Jones has some stronger teams in LA and good teams might also eat into your production (and again I’ve alluded to it multiple times, can’t find Ben/elgee/Thinking Basketball on Middleton with GA off – am confident it was a thing but can’t find it - take that as you will).
(note: have found this – not the same source but you can get to – for instance 2020 – where to my reading it says 641 minutes with GA off (may be some garbage time filters as before entering GA off the minutes are slightly less than full season) he was at 34.43 usage and 60.71 TS% - you’ll need to change the year, but maybe play around and see what other years look like https://www.fantasylabs.com/nba/on-off/?Source=4&PerMode=3&di349=2022-23&di1623=Regular&di306=Milwaukee%20Bucks&di1621=12485|12240|12268&FullGames=0&CurrentRosters=0&CurrentResults=0&MinPoss=1 )

Overall I’m not that bothered on the rank, it depends on one’s process, I’m not that focussed on individual years and if we’re tying in to one specific year then that playoff slump will hurt. As I said they’re all good players. Just felt a little dismissive so brought up his long term RAPM ranks and box peak because … idk, it felt as though he wasn’t in their league and yeah at first glance, without a deep dive on his all time position etc I didn’t see something like a gulf that seemed implied off, say, Peja.

And I may have overstated things wrt to Peja.....

Maybe I'm not critical enough of his defense, and [like Middleton in the year in question] he did lay a bit of an egg in the playoffs. So things could be substantially closer there than I implied on first pass.

I would argue none of them. They all have holes in their game, whether defense, shooting, shot creating, ball handling, etc. Haven't looked at the the metrics, but Khris is a much more complete player. Dumars may be the caveat there.