RealGM 2023 Top 100 Project - #29 (James Harden)

[iggymcfrack]

Vote: Harden
Alternate: Scottie Pippen
Nomination: Reggie Miller

Harden’s RS+PS impact signals look good:

These RAPM numbers you show in support of Harden represent Stockton’s 11th, 12th, 13th, 14th, 16th, and 17th best seasons by BPM. And yet Stockton ranks 17 spots ahead of Harden due to his incredible defense. Harden’s only played 14 seasons in total. You don’t think Stock deserves at least an alternate vote?

[f4p]

I guess the main things concerning me with Harden are (1) that his box numbers seem to trump his offensive impact & case as a floor raiser and

hmm, what makes you say that? he has a dominant ORAPM in the JE 97-22 (+6.5, basically in the group with jordan/steph behind lebron). and in terms of just teams:

Harden 2013-2021: +4.1 rORtg (1 of 9 seasons with KD)
Steph 2013-2021 (no 2020): +3.9 rORtg (3 of 8 seasons with KD)

and harden's teams win 2012/22/23 so it only goes more in his favor if i expand the range.

[OhayoKD]

Any chance we can get Stockton in this vote? I’d have him at least 16 spots ahead of any of the other nominees. Doesn’t seem like a great start to the voting for that outcome, but I feel like he has the best resume by a mile. I think his peak is underrated because people ignore what an absolute BEAST he was on the defensive end and he was Kareem-like in his later years.

I think the actual issue is that you ignore guard defense is way less valuable than any other sort of defense and signals that aren't built on artificial scales. No guard in nba history has approached being "kareem-like" defensively"

Spoiler:

https://forums.realgm.com/boards/viewtopic.php?f=64&t=2310981&hilit=in+1972&start=160

[trelos6]

Let’s agree that 29-33 will be the current group of nominees: Ewing, Harden, Pettit, Pippen, Stockton.

It brings up some interesting choices for picks 34-40.

I have Walt Frazier leading the list.

Then a whole host of names which I can see making it:

[iggymcfrack]

Any chance we can get Stockton in this vote? I’d have him at least 16 spots ahead of any of the other nominees. Doesn’t seem like a great start to the voting for that outcome, but I feel like he has the best resume by a mile. I think his peak is underrated because people ignore what an absolute BEAST he was on the defensive end and he was Kareem-like in his later years.

I think the actual issue is that you ignore guard defense is way less valuable than any other sort of defense and signals that aren't built on artificial scales. No guard in nba history has approached being "kareem-like" defensively"

Spoiler:

https://forums.realgm.com/boards/viewtopic.php?f=64&t=2310981&hilit=in+1972&start=160

No, no, I certainly wasn't trying to say he was Kareem-like in his defense specifically. Just in overall providing incredible value in his 13th through 19th seasons in the league. Obviously, an elite guard's defense won't be as valuable as an elite rim protector's, but that doesn't mean it's not valuable period. If Stockton's defense saves 2 points a game compared to an average defender, and Harden's defense gives up an extra point and a half compared to an average defender, that's a massive difference in the long run. 3.5 points a game is a bigger difference than the difference between the 1st place Nuggets and the 11th place Mavericks last year.

[falcolombardi]

Any chance we can get Stockton in this vote? I’d have him at least 16 spots ahead of any of the other nominees. Doesn’t seem like a great start to the voting for that outcome, but I feel like he has the best resume by a mile. I think his peak is underrated because people ignore what an absolute BEAST he was on the defensive end and he was Kareem-like in his later years.

I think the actual issue is that you ignore guard defense is way less valuable than any other sort of defense and signals that aren't built on artificial scales. No guard in nba history has approached being "kareem-like" defensively"

Spoiler:

https://forums.realgm.com/boards/viewtopic.php?f=64&t=2310981&hilit=in+1972&start=160

No, no, I certainly wasn't trying to say he was Kareem-like in his defense specifically. Just in overall providing incredible value in his 13th through 19th seasons in the league. Obviously, an elite guard's defense won't be as valuable as an elite rim protector's, but that doesn't mean it's not valuable period. If Stockton's defense saves 2 points a game compared to an average defender, and Harden's defense gives up an extra point and a half compared to an average defender, that's a massive difference in the long run. 3.5 points a game is a bigger difference than the difference between the 1st place Nuggets and the 11th place Mavericks last year.

I have serious doubts stockton D value is in the 2 points per game ballpark

Maybe 1 point is more of a realistic sell

[penbeast0]

Depends a lot on how much you value steals as to how valuable Stockton's defense was. Had a thread on here that used modern tracking stats to claim that steals were the single most valuable box score stat. Ie. 1 steal is worth more than 1 point, 1 rebound, 1 assist, 1 block, etc. That doesn't take into account the fact that some players gamble too much for steals so theirs are less valuable because they give it back a bit but Stockton wasn't that kind of thief.

Stockton (again) leads Kidd (again) on the all time steals chart by about 20% (again) just like he did in assists.

[iggymcfrack]

I think the actual issue is that you ignore guard defense is way less valuable than any other sort of defense and signals that aren't built on artificial scales. No guard in nba history has approached being "kareem-like" defensively"

Spoiler:

https://forums.realgm.com/boards/viewtopic.php?f=64&t=2310981&hilit=in+1972&start=160

No, no, I certainly wasn't trying to say he was Kareem-like in his defense specifically. Just in overall providing incredible value in his 13th through 19th seasons in the league. Obviously, an elite guard's defense won't be as valuable as an elite rim protector's, but that doesn't mean it's not valuable period. If Stockton's defense saves 2 points a game compared to an average defender, and Harden's defense gives up an extra point and a half compared to an average defender, that's a massive difference in the long run. 3.5 points a game is a bigger difference than the difference between the 1st place Nuggets and the 11th place Mavericks last year.

I have serious doubts stockton D value is in the 2 points per game ballpark

Maybe 1 point is more of a realistic sell

Even just the direct added value of his steals has to be worth more than that. He averaged 2.2 steals per game over the course of a 19 year career. That's more than any single player averaged in the NBA last season. Granted there are less steals overall nowadays, but at his peak, he averaged an extra 1.5 steals per game over an average point guard.

[penbeast0]

Vote: Stockton: Similar to Steve Nash offensively, better in structured offense, less outstanding when required to improvise, equivalent shooter on near equivalent volume, far superior defender, strong durability edge. Issues with playoff performances and a lack of rings are his biggest weakness.

Alternative Ewing: Strong defender, team leader, scorer, good durability, some playoff failures. Much as I root against the Knicks (and all NY teams), his excellent defense gives him extra value over the better offensive players like Barkley, Harden, and even Pettit. And, he could be a valuable 2nd or 3rd option where the great scorers frequently lose a lot of their value if asked to dial back their scoring to help the team.


Nominate Frazier: Efficient scorer, arguably GOAT guard defender, team leader, ran a widely admired (though not super efficient) offense, good playoffs, great finals. The two strikes against him are a relatively short prime/career, and being from the weaker 1970s era.

Atl Nominee: Pippen, though open to arguments otherwise. Another great two way player who gets dinged for not being a super volume scorer.

[HeartBreakKid]

It seems like era is penalizing Pettit more than Mikan, which is a little brutal on a guy with his resume. Were I voting, he'd have to be the guy here. A title, two scoring titles, 2 MVPs, 8 years as a top-5 MVP candidate in 11 seasons played (and another at 6th). Kept up his key traits even in his final season, into his early 30s with a heavy mileage log already behind him. Big regular season scorer with strong in-era efficiency, maintained reasonably well into the playoffs.

The resume suggests he should already be in several spots ago but playing in the 50s and early/mid 60s is clearly hurting him, despite other guys who have already gone in. Almost a lost star at this point. Triple threat action, a jump shot, great at drawing fouls, excellent rebounder. Dude had the whole package, and the accomplishments and accolades to boot.

Pettit's points, rebounds, assists, shooting efficiency all fell in the postseason and his WS/48 decline (0.213 -> 0.159) is the worst of any MVP-level player besides Malone and Embiid. He is lucky he played so long ago that most people never bothered to look at the numbers and he gets a pass for his playoffs dropoff.

Also, Cousy has a comparable resume to Pettit (1x MVP, 6x Champion, 10x All NBA First) and he likely won't even be top 50. It's not an era thing when players both right before and after them have already been voted in. The difference is playoffs performance.

This is an interesting point. I did notice they were down but not that much. Might have to retract my primary vote for him at the very least.

[rk2023]

I guess the main things concerning me with Harden are (1) that his box numbers seem to trump his offensive impact & case as a floor raiser and

hmm, what makes you say that? he has a dominant ORAPM in the JE 97-22 (+6.5, basically in the group with jordan/steph behind lebron). and in terms of just teams:

Harden 2013-2021: +4.1 rORtg (1 of 9 seasons with KD)
Steph 2013-2021 (no 2020): +3.9 rORtg (3 of 8 seasons with KD)

and harden's teams win 2012/22/23 so it only goes more in his favor if i expand the range.

Well, FWIW:

13-16 & 21-23 Steph (non KD years):
- 115.1 ORtg on to 105.3 ORtg off in RS
- 113.0 ORtg on to 107.4 ORtg off in PS

13-20 Harden (the platform won't let me do aggregate net-swing for >1 team ):
- 114.0 ORtg on to 108.9 ORtg off in RS
- 110.4 ORtg on to 105.9 ORtg off in PS

13-21 Steph (as you cited):
- 116.7 ORtg on to 105.8 ORtg off in RS
- 115.6 ORtg on to 108.3 ORtg off in PS

https://www.pbpstats.com/wowy-combos/nba
***All of this using PBPStats and filtering out low leverage minutes from the sample.

As for why I feel that way, Harden doesn't seem to have the offensive lift track record (eg. +/- and on/off) compared to others whom are functionally similar at a high-level w.r.t. similar RS statistical & box profiles (eg. LBJ, Wade, Jordan).

[MyUniBroDavis]

No, no, I certainly wasn't trying to say he was Kareem-like in his defense specifically. Just in overall providing incredible value in his 13th through 19th seasons in the league. Obviously, an elite guard's defense won't be as valuable as an elite rim protector's, but that doesn't mean it's not valuable period. If Stockton's defense saves 2 points a game compared to an average defender, and Harden's defense gives up an extra point and a half compared to an average defender, that's a massive difference in the long run. 3.5 points a game is a bigger difference than the difference between the 1st place Nuggets and the 11th place Mavericks last year.

I have serious doubts stockton D value is in the 2 points per game ballpark

Maybe 1 point is more of a realistic sell

Even just the direct added value of his steals has to be worth more than that. He averaged 2.2 steals per game over the course of a 19 year career. That's more than any single player averaged in the NBA last season. Granted there are less steals overall nowadays, but at his peak, he averaged an extra 1.5 steals per game over an average point guard.

That is like aggressively not how this works lol

[penbeast0]

I think the reason Pettit gets a playoff pass is the Hawks' ring in 1958. It's hard to call a guy a choker when in the close out game of the championship against the 11 titles in 13 years Celtics, he puts his team on his back and carried them to a 1 point victory with 50 points, 19 rebounds, and 19 of those points in the 4th quarter including 3 times in the last 6 minutes answering a Celtic bucket that brought them back to within 1 point when every single non-Hawks fan in the USA believed the Celtics would find a way to pull it out. (Whew, long sentence, let me take a breath.) Once you establish your "clutch" credentials like that, mere averages from the rest of your career has a hard time overcoming the narrative.

[penbeast0]

That is like aggressively not how this works lol

If steals truly have this value, why not?

[OhayoKD]

Depends a lot on how much you value steals as to how valuable Stockton's defense was. Had a thread on here that used modern tracking stats to claim that steals were the single most valuable box score stat. Ie. 1 steal is worth more than 1 point, 1 rebound, 1 assist, 1 block, etc. That doesn't take into account the fact that some players gamble too much for steals so theirs are less valuable because they give it back a bit but Stockton wasn't that kind of thief.

Stockton (again) leads Kidd (again) on the all time steals chart by about 20% (again) just like he did in assists.

Being the most valuable box-stat individually only really says it's strong relative to other box-stats. Overall team-steal totals do not correlate well with team defensive quality. That was the finding of the ucla study on the matter(there was slightly more offensive correlaition but not much) and per the 538 people who ran that irreplacibility study(though they never cited what specifically)

It's also evident if you go to bbr and just look at which teams have the most steals and cross-check that with which teams have the best defensive-rating.

Even without the drawback, they simply don't happen enough to move the needle like that and are probably more valuable offensively anyway(which gets us into double-counting issues)

No, no, I certainly wasn't trying to say he was Kareem-like in his defense specifically. Just in overall providing incredible value in his 13th through 19th seasons in the league. Obviously, an elite guard's defense won't be as valuable as an elite rim protector's, but that doesn't mean it's not valuable period. If Stockton's defense saves 2 points a game compared to an average defender, and Harden's defense gives up an extra point and a half compared to an average defender, that's a massive difference in the long run. 3.5 points a game is a bigger difference than the difference between the 1st place Nuggets and the 11th place Mavericks last year.

I have serious doubts stockton D value is in the 2 points per game ballpark

Maybe 1 point is more of a realistic sell

Even just the direct added value of his steals has to be worth more than that. He averaged 2.2 steals per game over the course of a 19 year career. That's more than any single player averaged in the NBA last season. Granted there are less steals overall nowadays, but at his peak, he averaged an extra 1.5 steals per game over an average point guard.

Nope. If that was how it worked, guards would not be the least impactful defenders

A team can get a defensive stop in scenarios where a steal occured and the process of searching for steals bleeds some value inofitself even if you're a more conservative thief. The main value of a steal is it generates an offensive possession where your team is at an advantage, somethign that would already largely be baked into stockton's assist and points.

[trelos6]

Vote: Stockton: Similar to Steve Nash offensively, better in structured offense, less outstanding when required to improvise, equivalent shooter on near equivalent volume, far superior defender, strong durability edge. Issues with playoff performances and a lack of rings are his biggest weakness.

Alternative Ewing: Strong defender, team leader, scorer, good durability, some playoff failures. Much as I root against the Knicks (and all NY teams), his excellent defense gives him extra value over the better offensive players like Barkley, Harden, and even Pettit. And, he could be a valuable 2nd or 3rd option where the great scorers frequently lose a lot of their value if asked to dial back their scoring to help the team.


Nominate Frazier: Efficient scorer, arguably GOAT guard defender, team leader, ran a widely admired (though not super efficient) offense, good playoffs, great finals. The two strikes against him are a relatively short prime/career, and being from the weaker 1970s era.

Atl Nominee: Pippen, though open to arguments otherwise. Another great two way player who gets dinged for not being a super volume scorer.

Pippen already nominated.

[OhayoKD]

I think the reason Pettit gets a playoff pass is the Hawks' ring in 1958. It's hard to call a guy a choker when in the close out game of the championship against the 11 titles in 13 years Celtics, he puts his team on his back and carried them to a 1 point victory with 50 points, 19 rebounds, and 19 of those points in the 4th quarter including 3 times in the last 6 minutes answering a Celtic bucket that brought them back to within 1 point when every single non-Hawks fan in the USA believed the Celtics would find a way to pull it out. (Whew, long sentence, let me take a breath.) Once you establish your "clutch" credentials like that, mere averages from the rest of your career has a hard time overcoming the narrative.

The caveat is that russell played only half that series and the celtics won the surrounding ones when russell played all the games(including one where the hawks had a 2-1 advantage).

To my knowledge no other russell opponent had the benefit of russell missing games.

[rk2023]

Let’s agree that 29-33 will be the current group of nominees: Ewing, Harden, Pettit, Pippen, Stockton.

It brings up some interesting choices for picks 34-40.

I have Walt Frazier leading the list.

Then a whole host of names which I can see making it (In no particular order):

Dwight Howard, Joel Embiid, Ben Wallace, Anthony Davis, Kawhi Leonard, Dikembe Mutombo, Artis Gimore, Jason Kidd, Reggie Miller, Clyde Drexler, Rick Barry, Dolph Schayes, John Havlicek, Elgin Baylor

Frazier certainly would be my next up here for a nominee after Miller and Kawhi. After that, it gets fun. My first guess is that I would have Baylor and Hondo the highest - but I certainly would have to relook into more unique cases like Kidd, Gilmore, Barry.

[MyUniBroDavis]

That is like aggressively not how this works lol

If steals truly have this value, why not?

Steals being the best box score stats in regards to defensive impact relative to other metrics does not equate to steal = 1 defensive point

Being the best part of an overall model that tends to not do nearly as well as it’s counterpart (offense) untill you ad impact data, should not be extrapolated to this single predictor within this model must have a lot of strength to it and have this X direct impact isolated alone, even the coefficient doesnt equate to a Steal = 1 defensive point either, and you have to also figure out if that impact is uniform or not the further you go away.

[Doctor MJ]

As none of my picks last time (Pettit, Pippen, Frazier, Miller), I expect I'll repeat my vote, but I find myself thinking that we don't seem to be spending a lot of time thinking about John Havlicek. I'm obviously not prepared to make the argument, but I wonder if I'm underrating him at the moment.