Iverson (and Dikembe) and the '01 Sixers---Gen Discussion (transplant)

[trex_8063]

Maybe thread-worthy topic that came up in the '24 POY Discussion Thread. Transplanting here to not further derail there.....

I feel [and this may not apply to you, Chuck.....you'll have to utilize self-assessment to determine for yourself] that there's an over-simplification of thinking that often occurs, along the lines of: "they win primarily because of their defense, and this is the best/most valuable DEFENSIVE player.....therefore he's clearly the best/most valuable guy on the team."

We see at least the fringes of this thinking every time someone wants to criticize Allen Iverson: they bring up that the '01 Sixers were good primarily because of their defense (which, fwiw, wasn't actually the case during the first 2-3 rounds of the playoffs: it was their OFFENSE that overperformed while their defense UNDERperformed, which carried them to the Finals ), and Iverson was one of the weaker defensive players in their primary rotation.
I've even seen person(s) argue that Dikembe/Ratliff was the best player on that team.

Couple things [opinions] I have regarding this line of thinking....
I feel one **Star Offensive** player can "carry" the offense (amid midling to poor help) better than one **Star Defensive** player can "carry" the defense (amid midling to poor help).

Think of:
*TMac on the '03 Magic-->that was actually the 10th-rated offense, despite the highest-minute supporting cast players being [in descending order of minutes]: Pat Garrity, aging Darrell Armstrong, Mike Miller, Jacque Vaughn, bloated Shawn Kemp, Andrew Declercq, Jeryl Sasser, and Gordon Giricek.
**Kobe on the '06 Lakers--->8th-rated offense with the following cast [again descending order of minutes]: Lamar Odom, Smush Parker, Kwame Brown, Devean George, Chris Mihm, Brian Cook, Sasha Vujacic, and Luke Walton.
Or.....
***Iverson on the '01 Sixers--->+0.6 rORTG 13th-rated offense [that way overperformed in the playoffs] with: Dikembe Mutombo/Theo Ratliff combo [Deke being the one around in playoffs], George Lynch, Aaron McKie, Tyrone Hill, Eric Snow, Toni Kukoc [though not for playoffs], Jumaine Jones, Kevin Ollie.


Vs (here going with "defensive stars" with BAD defensive casts):
*Draymond Green on the '20 Warriors--->26th-rated defense. I know he missed a lot of time, but even when he was on-court [note: not just "games he was present for", but when he was literally ON the court] their rDRTG was still +1.6 (would have been tied for 20th in the league). Cast was: Eric Pacshall, Glenn Robinson III, Damian Lee, Alec Burkes, Jordan Poole, Marques Chriss, D'Angelo Russell, Ky Bowman, Willie Cauley-Stein, and Omari Spellman.
**Dikembe Mutombo on the '00 Hawks--->tied for 25th [of 29] teams defensively with this cast: Alan Henderson, Jim Jackson, Isaiah Rider, Bimbo Coles, Jason Terry, LaPhonso Ellis, Lorenzen Wright, Roshown McLeod. Deke played a team-high nearly 3k minutes on that team, too.
***Nate Thurmond on the '74 Warriors--->+0.7 rDRTG, with Thurmond being a close 4th in total minutes. Cast was: Rick Barry, Cazzie Russell, Jeff Mullins, Butch Beard, Jim Barnett, Clyde Lee, *George Johnson, Charles Johnson, and Derrek Dickey. (*GJ was actually very good defensively, nor would I describe everyone listed here as flatly "bad" defensively; and yet...)


The reason for this phenomenon is [imo], that a single great defender can only clean up so many messes that the rest of the team creates. To his credit, a guy like Gobert can erase a number of sins. But to a greater degree, excellent TEAM defense is the product of a solid coordinated TEAM effort (especially in TODAY'S league).

Solid offense often relies upon solid team contribution, too; but an offensive star-level player can usually generate a decent(ish) look for someone even when everything else breaks down. And even in today's league, there are still A LOT of offensive possessions that come down to some iso ball action. Surprisingly(ish) (only because they're one of the best offenses in the league), you still see a lot of iso action on the current Celtics team (usually Tatum or Brown) . The Celtics are just able to compliment that with a plethora of shooters and guys who take really good care of the ball.
Lot of iso on the Suns this year, too.


So circling back to Gobert and Edwards on the TWolves:
Yes, Gobert is the most important defensive piece.......but that's actually still a pretty good defensive squad even without him. I know it's a 1-game sample, but look to game 2 against the Nuggets as evidence. That 1st half was possible the single-greatest team defensive performance I've ever seen; and Rudy wasn't even dressed!
McDaniels and NAW are both long, athletic, super-versatile perimeter defenders. KAT, when he's bought-in defensively, is a solid low-post defender. Naz provides some rim protection. Kyle Anderson is a savvy veteran defender; ditto Mike Conley (though small). And Edwards himself is a very respectable perimeter defender (so he's a significantly part of that defensive rating).

There almost isn't a true weak spot [on defense] amid their primary line-up.

On offense, they're far from "crap"; I don't want to imply that Edwards has "carried" the offense all season. But if you replaced him with, say......Reggie Jackson; that would have been a significantly below average offense, and likely would struggle MIGHTILY in the playoffs.
But because Edwards is who he is, this was actually a +0.3 rORTG team in the rs that performed as a monstrous +13.5 rORTG [relative to defense faced] in the 1st round, and a +1.6 rORTG in the WCSF.


Erase from existence EITHER player, and I'm sure the TWolves are not in the WCF. But I just don't think it's as simple as noting that they're largely dominant because of their defense and that Rudy is the best defensive player.

So I agree with the gist, that one player is not the entire team so so can be, for instance a good offensive player on a poor offensive team. Ditto that it isn't as simple as "noting that they're largely dominant because of their defense and that Rudy is the best defensive player."

Regarding 76ers best player I think one would have to be clear what window one is judging on.

With regard to overperformed and underperformed on each end in a particular window of the playoffs ... I would note two things
1) if we're talking about the cause of winning - I would be wary of, or want clarity on what "overperformed" and "underperformed" means. If it's versus RS expectations and my team was expected to be the GOAT defense level and poor offensively and we're merely elite and average respectively ... the defense is still the thing driving wins. I don't think you should punish players or teams or "ends" for great regular seasons.
2) we're getting quite abstract from player contribution. Fwiw, and tiny off samples, very noisy measure etc, if it's regarding point margin and that as an indication of driving goodness his playoff on-off is 5th out of 7 rotation player and clearly behind Mutombo (his "on" is also 5th).

I think there is something to a great offensive player being able to do more by themself in you decide where the ball goes on offense (unless the defense really sell-out to keep it away from a star, at a significant cost), where a great defender can't control where the team is attacked. It seems plausible.

At the same time ... there are limitations it depends on context. It may be that a porous exterior defender makes having ... say ... a Shawn Bradley as the backline (and rim protectors can in this way cover multiple players and their highest value shots). And the 1 man offense ... at the pro level ... I think there's a ceiling there.

Regarding specific seasons as seasons...
- 2000 Mutombo is +11.6 on-off, he's impactful, and presumably on defense ... it's just starting from a really low baseline.
- '74 Thurmond might be regarded to be post prime, has a good backup and he misses a chunk of the season which (1) ... probably makes their number worse and (2) per ElGee's spreadsheet still suggested some, not huge but not nothing impact (2.7 SRS change).

Regarding specific seasons as evidence maybe just meant as illustration but even if "true" (see doubts above) it's rather ad hoc.

To Gobert the RS defensive impact signal isn't close to as night and day as it may have been in the past (off, for instance https://www.cleaningtheglass.com/stats/player/1310/onoff#tab-team_efficiency). But then the player in the particular comp ... had a ... nice? ... RS year. PER just short of 20, BPM just short of 3.5, WS/48 at .130 (where, unlike BPM the greater value is credited on the defensive side) ... He's had a very good first two rounds but I don't know ... for PoY that's quite a small sample and fwiw, it felt like you weren't that enthused about '07 Baron Davis over a similar sample with (on average of the 3 Reference composites, though varying depending on preference) a clearly better box-side output and fwiw (huge caveats regarding samples, but if the playoff progression and playoffs is that important ... this is important context to that) better playoff side impact indicators (both in raw terms and within rank on team for on-off).

So I agree with the gist, that one player is not the entire team so so can be, for instance a good offensive player on a poor offensive team. Ditto that it isn't as simple as "noting that they're largely dominant because of their defense and that Rudy is the best defensive player."

Regarding 76ers best player I think one would have to be clear what window one is judging on.

With regard to overperformed and underperformed on each end in a particular window of the playoffs ... I would note two things
1) if we're talking about the cause of winning - I would be wary of, or want clarity on what "overperformed" and "underperformed" means. If it's versus RS expectations and my team was expected to be the GOAT defense level and poor offensively and we're merely elite and average respectively ... the defense is still the thing driving wins. I don't think you should punish players or teams or "ends" for great regular seasons.

Perhaps I should have been more clear: the '01 Sixer offense significantly outperformed their defense in the playoffs. I don't mean that in a "vs RS expectations" sense; I mean that in an absolute sense.

1st round (relative to ORtg/DRtg being faced): Offense outperformed the defense by +4.2 rating (they were a +5.6 rORTG, -1.4 rDRTG [remember negative is good for rDRTG]).
ECSF: Offense outperformed the defense by +7.5 (+5.4 rORTG, +2.1 rDRTG).
ECF: This round was basically a wash [marginaly edge to defense in absolute sense], with the defense outperforming the offense by a mere +0.3 rating (+2.1 rORTG, -2.4 rDRTG). Considering this was a team built for defense, and looking at their rs standard, this was still a small underperformance for the D, and a small overperformance for the O (but yes: in absolute terms the defense "won" by the narrowest of margins).

So I would stand by a statement suggesting they were propelled to the Finals by way of their offense rising to the occasion, while their defense [particularly relative to cast/roster make-up and rs standard] was a bit of a shrinking violet.

fwiw, in the Finals, the defense again "wins", but only by 0.6, with BOTH underperforming (-2.2 rORTG, +1.6 rDRTG). Remember the '01 Lakers went super-nova in the ps.

Don't know what to make of the playoff on/off for Iverson on that team, given he's only "off" a grand total of 87 minutes in the entire playoffs (he did have the highest offensive on/off, fwiw, but again the sample....).

Regarding specific seasons as seasons...
- 2000 Mutombo is +11.6 on-off, he's impactful, and presumably on defense ... it's just starting from a really low baseline.

On the '03 Magic, TMac had a +13.0 on/off, and more specifically a +17.5 offensive on/off. When he was off-court, that cast I noted was an astounding -11.8 rORTG (how's that for a low baseline?). It would have been worst in the league, despite that fact that in actuality there was an atrocious Nuggets team with a -11.4 rORTG. With TMac on the court, though, they improved to +5.7 rORTG (would have ben 2nd in the league). He appears to literally almost lift that cast from historically bad worst in the league offense to best in the league.

Kobe Bryant did basically the same with the '06 Laker cast I mentioned, posting a ridiculous +18.9 offensive on/off.


I don't think there's an instance (in the databall era) where we see a single defender having that kind of impact within their side of the ball.

On the former thanks for clarification.

Not databall and not official stats and slightly smaller number (and smaller on [but greater off]) but the spreadsheets that estimated this for the 76ers off their +/- data it had '92 Bol's defense at -17.1. That was the otoh likeliest candidate I had to be in that ballpark.

I would also say saying that sure the offensive guy's number is larger ... but Mutombo is someone you chose, and the team level 2000 data presented ... I might venture, if taken at face value might have led to an impression that Mutombo wasn't impactful and that seems far from the truth.

The other thing I'm curious about but only have time to briefly touch on here is to what extent we can expect a team to match/replicate their performance on a given end in a Bo5 or Bo7 sample.

Don't know how much it moves the needle and or to what extent it was a planning failure and fouling the wrong people ... but at a glance
Indy shot .827 from the stripe (.766 RS)
Toronto shot .782 (.747)
Milwaukee shot .842 (.787)

Don't know what to make of Iverson's on/off either. As I say very small samples. But the disparity with Mutombo suggests they happened to lose a lot of ground without him. But regardless if good team level defense doesn't just mean "defender", offense doesn't mean ... "offensive player". Their team FT% goes up a little too, even as Iverson's gets worse ...

[trex_8063]

Picking up where we left off......


It's a fair question to ask (the one re: FT% of opponents in that playoff run). I think such shifts affect the DRtg less with the '01 Sixers than it might with other teams, simply because they didn't put opponents to the line as much as other teams (3rd in the league in opp FTAr; and they consistently hit [roughly] their rs standard in that regard in each of the first three rounds, reducing the number of FTA for each opponent).

And in at least one of those rounds [1st round vs Pacers] it could be suggested that the Sixer defense is partly to blame for the improved FT%, on the basis of WHO they're putting to the line:
In the rs, Reggie Miller (~92-93% FT shooter that year) averaged 4.3 FTA/g and 17.3% of the TEAM'S total attempts; in that 1st round series he averaged 7.5 FTA/g, accounting for 40% of the team's attempts.

[Owly]

Picking up where we left off......


It's a fair question to ask (the one re: FT% of opponents in that playoff run). I think such shifts affect the DRtg less with the '01 Sixers than it might with other teams, simply because they didn't put opponents to the line as much as other teams (3rd in the league in opp FTAr; and they consistently hit [roughly] their rs standard in that regard in each of the first three rounds, reducing the number of FTA for each opponent).

And in at least one of those rounds [1st round vs Pacers] it could be suggested that the Sixer defense is partly to blame for the improved FT%, on the basis of WHO they're putting to the line:
In the rs, Reggie Miller (~92-93% FT shooter that year) averaged 4.3 FTA/g and 17.3% of the TEAM'S total attempts; in that 1st round series he averaged 7.5 FTA/g, accounting for 40% of the team's attempts.

So this is getting a bit tangential but at a glance

I don't know how much difference FT% makes. It's true that if you foul less it matters less. That, as I'm sure you understand, doesn't alter the underlying point.

As to whether it is luck based on who was fouled and Miller I'll fight back a bit but generally add context and alternate perspectives) somewhat on a number of levels

1) having granted the possibility of who was fouled as a factor ... crediting/punishing that as defense rather than luck tends to regard foul as intentional rather than luck. I'd suggest most aren't and barring "Hack-a..." or end of game, those that are where you've already ceded an advantage. Knowing your cover will affect how you defend but ... IDK ... maybe there really are "smart" fouling team over the long term. My initial inclination is it's more likely luck.

2) Miller is only one player. Really one would need to do this systematically.
2a) But if you are doing this for great shooters ... Ray Allen, if FTr is just FTA/FGA which as I understand it Reference's version is (and it appears ot be on Miller's numbers), was more significantly reduced (or per above just happened to be fouled less) in volume

3) The framing of Miller's rate given is one angle ... another would be ... in somewhat of an outlier low among all seasons and surrounding seasons Miller is at .296 FTr that regular season. It went up to .333 in that playoff i.e. that series. Given the small sample that's really not so different and is actually slightly below his RS norms of surrounding seasons. Then too Miller tended to up his usage in the playoffs and very marginally his FTr and FT% ... for those who think this is "real" more than noise, things like Miller even getting more shots (and we're tangenting further now) is arguably nothing to do with the 76ers and more to do with whether RS provides an accurate baseline for Miller.

4) Having used it because it illustrates that Miller wasn't fouled more out of nothing, he was shooting more (and he was playing more too) ... there are issues with FTr more generally depending what it's purporting to measure. Here you'd be adjudged to be fouling less if you gave more and-1s instead of getting a miss and a 2 (or 3) shot foul and forcing the miss. To some extent this could lead back to issues discussed in point one regarding intent and what the alternatives are.

And then to the extent this comes back to Iverson rather than offense/defense one could then look at how much and who he was fouling. But that would be awfully deep down a rabbit hole on stuff that might just be chance.

[penbeast0]

And, if Iverson is the cause of Philly playing better on offense in the playoffs as you say, strange how it takes place with him sporting a roughly equivalent usage and ast%, but being worse at ts%, foul draw, and turnover %. Maybe just a small sample size fluke? Mutombo does have a game score higher then Iverson's in the ECF, but not close in the other 3 series.

[tsherkin]

Some thoughts on Philly's playoff O that year...

RS:

103.6 ORTG (13th), 90.6 pace, 51.8% (14th, spot-on league average), 13th in 2FG at 46.4% (+0.010%), 4th in FTA, 5th in ORB, 19th in eFG%, 21st in TOV, 2nd in ORB% (31.2%), 4th in FT/FGA (.262), 3rd in FTr (.351). 79.1 FGA/g

PS:

105.7 ORTG (6th), 87.0 pace, .332 FTr (7th), .498% TS (11th), 44.5% eFG (-0.07%, 11th), 11.8% TOV (4th), 32.7% ORB (2nd), .253 FT/FGA (8th). 80.8 FGA/g.

So they shot worse, drew fouls less effectively, shot the ball more at lower efficiency... but they turned the ball over less and hit the offensive boards more effectively (which was already a strength).

So in the strictest sense, they were actually worse at scoring.

Iverson himself was a 25.5 FGA/g guy on 44.7% eFG, .397 FTr, 51.8% TS (league average) and 10.0% TOV in the RS.

In the PS, he was a 30.0 FGA/g on 42.5% eFG, .315 FTr, 48.0% TS (-3.1% rTS compared to playoff average) and 7.7% TOV.

His USG moved from 35.9% to 36.8% and he went from 23.0% AST to 29.1%.

Just to land some details. He took on a slightly elevated usage rate, shot far worse, but was dragging up their FTr and dragging down their TOV% while the team overall did a better job around him on the offensive glass. And they largely got away with it due to conference competition.

So, in essence, because Philly was so gorram useless at everything but offensive rebounding, Iverson shooting more at lower percentage and what was still a good FTr while limiting his turnovers by jacking up bricks all game long actually did help their overall offense, because the possessions used by the rest of the team on O weren't any good.