What would Duncan have had to do to be seen in the same category as Kareem?

[drza]

...

What makes you believe that Shaq was a better offensive player than Kareem?

I'll start with some per-100 stats, even though that's not where my argument rests. But, it's good to have some numbers to reference in addition to the more description-based discussions below. So:

Kareem per-100 stats 1974: 29.9 PP100, 59.2 TS%, 3.1 ORBP100, 4.5 AP100, 2.7 TO
22.0 FGA/100, 7.1 FTA/100 = ~25.1 possessions used on shots

Shaq per-100 stats career: 35.2 PP100, 58.6 TS%, 5.2 ORBP100, 3.7 AP100, 4.1 TO
24.0 FGA/100, 13.9 FTA/100 = ~30.2 possessions used on shots

Next, I want to post a well-known quote from Bill Russell:

If you take all the seconds added up shooting and rebounding, it comes to about three minutes. Now, out of a 48-minute game, three minutes are concerned with shooting and rebounding. What is going on the other 45 minutes?

The Russell quote shapes more of my thought process here than the numbers do, and hopefully give a frame of reference for how I see things. Because, if we assume roughly 100 possessions (not perfect, but round numbers), on both offense and defense, then what the box score numbers I posted tells us is that Kareem scored about 30 points on about 25 shooting possessions, while facilitating another 9ish points on 7ish passing possessions. Shaq generated about 5 more points on 5 more shooting possessions, and about 1.5 points less on 1 more passing possession.

Kareem's 39ish points on 32ish offensive possessions used, vs Shaq's 42.5ish points on 38ish possessions. I mean, we could debate which would be more valuable to a team. You could say Kareem was a bit more efficient on the whole, but Shaq was producing higher volume more directly as the scorer...but to me, the end result of these (average) 35ish possessions could get lost a bit in the noise, because, in the vein of Russell's quote...

What's going on in the other 65ish offensive possessions on offense? Or the 100 possessions on defense?

Since we're focusing only on the offensive end for this comp, let's forget about the 100 defensive possessions for now. On offense alone, we're still talking about 2/3 of the time that the bigs are on the court without finishing the possession.

One of the advancements basketball analysis that has fallen out of the longitudinal +/- studies is a series of descriptions of elements of the game that aren't measured in the box scores. Because, at least anecdotally, there doesn't seem to be a huge correlation between high volume/high efficiency finishing and the best offensive impacts. Instead, we've started using terms like "defense warping", "spacing" and (team) offense-creating to help explain what we tend to see. And, again anecdotally, these terms DO tend to correlate with the best offensive impacts.

Bringing it back to Kareem vs Shaq
While Kareem was the most dominant big of his time, I think Shaq is the undisputed GOAT at warping opposing defenses. Opponents would play extra non-skilled bigs to use as fodder to try to slow him down (with 6 fouls, if nothing else), thus limiting their own offensive upside. They'd shade the entire team toward Shaq even when he didn't have the ball, then if he did get it he'd get routinely tripled in the post. Kenny Smith used to always say that Kobe Bryant was "the best 1-on-1 player in the world, that actually gets to play 1-on-1" because, as great as Kobe was, defenses had to overload on Shaq.

As much attention as Kareem drew, and he drew plenty, I think Shaq beats him handily in this aspect of the game. And by drawing so much defensive attention, he was creating better looks for his teammates simply by being on the court...to a higher degree than Kareem (or anyone). And he was doing this for all 65 of the possessions that he wasn't finishing with a shot, assist or TO.

A few years back, Kirk Goldsberry presented a decision-making stat, where he used advanced scouting data to generate a decision-matrix for the ball-handler in every situation. It utilized a fractional approach, where adding up the fractional increases/decreases in a team's probability to score based on the decision-making of the ball-handler produced a quantified way to estimate how much such decision-making could add up over the course of a game.

Similarly, I'd argue that Shaq's ability to warp a defense, multiple times, on every possession, over a full game, added up to a larger positive change in his team's offensive fortunes than any small differences one way or the other in the boxscore stats for the possessions that both actually used.

TL;DR: I argue that Shaq had more offensive impact than Kareem, not because he scored at a higher rate on similar efficiency, but because his overall offensive package (including and especially his ability to warp defenses) created more offensive opportunities for his team over the course of a game. And, therefore, Shaq's presence on the court would lead to a larger improvement in his team's offensive scoring margins than Kareem's would.

[falcolombardi]

...

What makes you believe that Shaq was a better offensive player than Kareem?

I'll start with some per-100 stats, even though that's not where my argument rests. But, it's good to have some numbers to reference in addition to the more description-based discussions below. So:

Kareem per-100 stats 1974: 29.9 PP100, 59.2 TS%, 3.1 ORBP100, 4.5 AP100, 2.7 TO
22.0 FGA/100, 7.1 FTA/100 = ~25.1 possessions used on shots

Shaq per-100 stats career: 35.2 PP100, 58.6 TS%, 5.2 ORBP100, 3.7 AP100, 4.1 TO
24.0 FGA/100, 13.9 FTA/100 = ~30.2 possessions used on shots

Next, I want to post a well-known quote from Bill Russell:

If you take all the seconds added up shooting and rebounding, it comes to about three minutes. Now, out of a 48-minute game, three minutes are concerned with shooting and rebounding. What is going on the other 45 minutes?

The Russell quote shapes more of my thought process here than the numbers do, and hopefully give a frame of reference for how I see things. Because, if we assume roughly 100 possessions (not perfect, but round numbers), on both offense and defense, then what the box score numbers I posted tells us is that Kareem scored about 30 points on about 25 shooting possessions, while facilitating another 9ish points on 7ish passing possessions. Shaq generated about 5 more points on 5 more shooting possessions, and about 1.5 points less on 1 more passing possession.

Kareem's 39ish points on 32ish offensive possessions used, vs Shaq's 42.5ish points on 38ish possessions. I mean, we could debate which would be more valuable to a team. You could say Kareem was a bit more efficient on the whole, but Shaq was producing higher volume more directly as the scorer...but to me, the end result of these (average) 35ish possessions could get lost a bit in the noise, because, in the vein of Russell's quote...

What's going on in the other 65ish offensive possessions on offense? Or the 100 possessions on defense?

Since we're focusing only on the offensive end for this comp, let's forget about the 100 defensive possessions for now. On offense alone, we're still talking about 2/3 of the time that the bigs are on the court without finishing the possession.

One of the advancements basketball analysis that has fallen out of the longitudinal +/- studies is a series of descriptions of elements of the game that aren't measured in the box scores. Because, at least anecdotally, there doesn't seem to be a huge correlation between high volume/high efficiency finishing and the best offensive impacts. Instead, we've started using terms like "defense warping", "spacing" and (team) offense-creating to help explain what we tend to see. And, again anecdotally, these terms DO tend to correlate with the best offensive impacts.

Bringing it back to Kareem vs Shaq
While Kareem was the most dominant big of his time, I think Shaq is the undisputed GOAT at warping opposing defenses. Opponents would play extra non-skilled bigs to use as fodder to try to slow him down (with 6 fouls, if nothing else), thus limiting their own offensive upside. They'd shade the entire team toward Shaq even when he didn't have the ball, then if he did get it he'd get routinely tripled in the post. Kenny Smith used to always say that Kobe Bryant was "the best 1-on-1 player in the world, that actually gets to play 1-on-1" because, as great as Kobe was, defenses had to overload on Shaq.

As much attention as Kareem drew, and he drew plenty, I think Shaq beats him handily in this aspect of the game. And by drawing so much defensive attention, he was creating better looks for his teammates simply by being on the court...to a higher degree than Kareem (or anyone). And he was doing this for all 65 of the possessions that he wasn't finishing with a shot, assist or TO.

A few years back, Kirk Goldsberry presented a decision-making stat, where he used advanced scouting data to generate a decision-matrix for the ball-handler in every situation. It utilized a fractional approach, where adding up the fractional increases/decreases in a team's probability to score based on the decision-making of the ball-handler produced a quantified way to estimate how much such decision-making could add up over the course of a game.

Similarly, I'd argue that Shaq's ability to warp a defense, multiple times, on every possession, over a full game, added up to a larger positive change in his team's offensive fortunes than any small differences one way or the other in the boxscore stats for the possessions that both actually used.

TL;DR: I argue that Shaq had more offensive impact than Kareem, not because he scored at a higher rate on similar efficiency, but because his overall offensive package (including and especially his ability to warp defenses) created more offensive opportunities for his team over the course of a game. And, therefore, Shaq's presence on the court would lead to a larger improvement in his team's offensive scoring margins than Kareem's would.

Doesnt this fail to account for the difference in league average efficiency between their peaks ? Peak shaq played in a league with illegal defense rules AND a 3 point line that was starting to be used a bit if mainly as a lower volume spacing tool

The difference in their scoring is way more than 1%. Kareem ts add relative to league average is in a league of its own compared to shaq for this reason. Kareem added more points relative to league average with his smaller volume than shaq did on his

I think a raw volume and raw efficiency comparision is misleading for this reason. Kareem just added more points with his scoring volume/efficiency combo than shaq

I remember watching kareem vs walton in 77 very closely and kareem had a huge warping effect too, bill walton would preemptively front him far from the hoop (thus creating driving lanes) to prevent kareem from catching and effortlessly scoring on him

One of the greatest rim protectors ever would go out of his way to front another big outside the paint snd leave the paint open. This in a era without 3 point line where bigs were hardly pulled out of the paint due to the marginal value of the midrange

Other times blazers would imediately double or triple kareem as soon as he got the ball to prevent him from shooting even if he could pass to a open jumper

And whenever walton played him straight up kareem would score on him like he was a prop.

And the point about possesions without the ball applies even more to defense than it does to offense, kareem seems like easily the more active (compared to shaq) and quick defensive player who protects the rim without being as exploited outside it

[tsherkin]

Want to push back a bit on this. When comparing Kareem and Duncan as scorers, you note that their per 100 possession scoring is similar, but of course Duncan didn't have as many minutes. I think the problem with this thinking is that I highly doubt Duncan could have maintained his scoring averages in more minutes, while also maintaining his rebounding and defense.

That was specifically my point, yes.

[drza]

Spoiler:

Want to push back a bit on this. When comparing Kareem and Duncan as scorers, you note that their per 100 possession scoring is similar, but of course Duncan didn't have as many minutes. I think the problem with this thinking is that I highly doubt Duncan could have maintained his scoring averages in more minutes, while also maintaining his rebounding and defense.

I think this is one aspect about impact and players that rarely gets talked about which is that they rarely actually do everything they're capable of on the court all the time. Bigs have more limited stamina (especially as they age) and have to pick and choose there spots on either side of the ball. Kareem in the 80s, especially in the 1st half, focused most of his energy on half-court offense at the expense of rebounding especially as well as defense. Duncan sacrificed minutes and scoring for rebounding and defense. As such, I don't think it makes much sense to try and equate Duncan and Kareem as scorers while simultaneously giving Duncan credit for his defense/rebounding as is.

If you want to argue that Duncan's longevity curve was better because bigs are inherently more valuable as defensive lynchpins specifically as they age, that's perfectly reasonable, but throwing in the scoring aspect of it is overkill IMO.

Moreover, I think that the idea Duncan was comparable to Kareem as a scorer or overall offensive engine is pretty questionable. 1st, the pace estimates underrate bigs like Kareem because he didn't really benefit from transition, and moreover when you factor in spacing differences and efficiency relative to the league, the gap grows wider. This is illustrated by the fact that Kareem's 1986 season was about equal to Duncan's 2003 season in terms of per75 relative TS%. Essentially, a 38 year old Kareem was as good a scorer as peak Duncan, which I think has as much to do with scoring becoming easier per possession over the course of Kareem's career as it does with his incredible longevity.

Spoiler:

Fair responses, based on what I wrote here. For me, my post was a bit of an extension to the conversation from a few weeks ago that I think indirectly led to this thread. In my last post in that thread, I think I touched on a lot of what you wrote here re: defensive impact more important to bigs than offensive impact, the scoring conversation being more tangential to my point than a key factor in it, necessity for some sort of pace adjustment despite the conversations on fast break/iso/scaling effects, etc. I was going to just paste the link to the post, but it's not super-long, so I'll also quote inset it here. https://forums.realgm.com/boards/viewtopic.php?p=100834900#p100834900

Spoiler:

Short answer: yes. Defensive impact, especially among big men, is often a more important factor to their team's fortunes than scoring. Kareem is one of only a handful of big men in history that really had an all-history impact as a score-first player (Kareem, Shaq, Dirk, arguably Jokic, maybe a few others). Typically, I'd much rather my big man be defensively dominant and good offensively as opposed to vice versa.

Plus, elephant in the room, if you're going to compare volume boxscore stats you have to, in some way, account for pace. I've been in all the arguments through the years about pace adjusting, and whether it's accurate for high-volume/iso scorers because there are questions about scaling iso vs pace, and all that. I get all that. But just, KISS, you can't IMO make any sort of reasonable volume-based comparison of what a player produces in 110 possessions per game vs what another produces in 90 possessions/game. It's not viable. So, I'm one that (if I'm going to factor in volume box score stats) will often utilize either the per-100 numbers, or sometimes I'll look at the ratio of each player's volume vs their team's volume to try to get them on a similar level.

Anyway. As quick datapoints, I'll see your 12,000 more total points in their careers and raise you their scoring rate per 100 possessions:

Kareem: 29.9 PP100 (1974 season till retirement; 16 seasons (all they have on B-R))
Duncan: 29.7 PP100 (entire 19 season career)

Now, again, I don't particularly care about their relative scoring volumes. I tend to evaluate by impact, and to the degree I've been able to determine, I'm comfortable that Duncan's impact was larger than Kareem's. With that said, and with the pace adjustment caveats of the above paragraph in place...I'm kind of surprised, myself, at how similar those numbers were. BECAUSE so much of Kareem's legacy in these GOAT talks is based on his role as the scoring king, while Duncan is often considered just a "good" scorer in a historic sense. But, it really is true that the era that Duncan played in was one of the slowest, most-difficult-to-amass-points eras in NBA history. Maybe pace adjusting isn't perfect, but I think it's conservatively fair to say that on the continuum of NBA history, Duncan was probably a much better relative scorer than his volume stats indicate.

Spoiler:

This is a conversation that could go a lot of different ways, and could spark some fun analysis/revelations. I've said a few times that I've found Duncan's impact to be larger than Kareem's, but there hasn't been much of that conversation in this thread. But we could start with a question about mechanisms:

Kareem is on the short list of elite scoring threats in history. Does that make him one of the elite offensive impact players in history? That'd be unusual for a scoring big, but not unheard of. But I haven't necessarily seen the evidence that he had an offensive impact analogous to, for example peak Shaq's, or even current Jokic. In the absence of that, I'd argue that Duncan's demonstrated ability to have all-history level impacts on defense would put him in the driver's seat in an initial comp between the two. Want to start there, and push back with some evidence that I'm wrong, and that Kareem's offense was more impactful and/or that Duncan's defense was less impactful than I'm suggesting here?

I do remember that thread and yea, it's an interesting conversation for sure that can go many directions. On the offensive impact point, comparing Kareem to say Shaq isn't close to being an apples to apples comparison, and this is where impact metrics have limitations. Kareem definitely didn't have the offensive impact Shaq had, in era, but the spacing in Kareem's era alone puts him at a massive disadvantage in that regard. Just looking at their respective skillsets, if you put Kareem in Shaq's era I think he would be very close to if not slightly better than Shaq offensively. While I know hypotheticals are prone to bias, I do think it's inarguable that Kareem would have significantly more impact on offense in Shaq's era than he had in his own, but having Shaq over him is perfectly valid of course.

There were also some interesting points brought up in another thread about how the Spurs defense didn't seem especially dependent on Duncan and how ideal a situation Duncan was brought up in. I can't pretend I've thoroughly studied Duncan's career enough to be able to say one way or the other, but it was an interesting counter-narrative to the general consensus that Duncan is a top-5ish level defender ever.

Kareem was an unstoppable force during his prime on offense.

Duncan was never a lock down defender, routinely was assigned the lesser big on defense and was freed up to cover defensively as a help defender. He never had a massive impact on any opposing big on defense that I could find.

Even with all of that. You would expect to see a significant impact on team defense based on Tim's presence. He joined a team that was 3rd defensively in the league the year before Robinson was out. SAS hung out in the 1-3 league DTRG rankings until 08-09 with Robinson and then Bowen being significant contributors on defense. Once Bowen's time dropped in 2009 they slid to 5thand then 8th, 11th, 10th until K Leonard became a starter and they jumped back into their 1-3 rankings for 4 years. Sure Duncan alone can't hold a team up on his shoulders defensively all by himself, but in 2017 without him SAS fell all the way to FIRST place defensively without him. They were first in 2016 as well even though he only played 1536 minutes.

Early in his career he was routinely putting in around 3200 minutes per season. Then in 2004 they reduced his minutes to around 2550. That's about 17 less games played at 39 MPG. Even with him missing that much time they continued to post league defensive rankings 1,1,1,2,3 then we hit Bowen leaving and drop to 5th.

SAS were a great defensive team that had a defensive culture before and after Duncan arrived.

I just don't see what measure anyone could hold up to suggest Duncan's defense was ATG compared to Kareem's offense. Duncan was absolutely a great defender, but he wasn't a stand out defender winning DPOY every year or in fact any year. (which is probably a crime in and of itself). The Tim love seems to do a lot of heavy lifting in discussions about him.

Lets not forget that Kareem was a fine defensive player in his own right

yeah, the falling off to 1st after duncan retired shows that no one was in a more career-long ideal situation than duncan. came in with a veteran team ready to win who had just added a GOAT coach, left with his best supporting cast, one that could win 67 games and have a 10+ net rating even with duncan playing 60 games and 25 mpg, and a team that could still win 61 and lead the league in defense after he left. i often wonder how much of the "longevity" argument for duncan is really just that he got the best supporting cast of his career at the end.

most guys are still trying to be the 20+ ppg scorers and/or defensive anchors they were as younger players while the supporting cast gradually falls off. we get to see them get older, less efficient, not be able to do it any more. we get to see them lose. duncan just got to skip that. it looked like he might experience that from 2008 to 2010, when the spurs were starting to look like a trio of aging hall of famers with a weaker supporting cast, which is why there were so many "are the spurs done?" conversations in those years. but then they added diaw and splitter and mills and green, and oh yeah, another hall of famer using a non-lottery pick. so now he could just show up on a random tuesday in january, put up 6 pts and 11 rebounds in 26 minutes, with 4 of those points coming off easy layups setup from parker/ginobili drives, while the spurs won by 20 and everyone would say how impactful he was. even though he was nowhere near the 25 ppg kind of guy he was at his peak who could carry everything. he could just play his ideal role with no pressure to step outside of it. take it easy on offense and focus on defense. could take off back to backs or nurse injuries for an extra game or two, instead of getting worn down like most greats. so his perfect situation just gets more perfect because he doesn't have to put up with the stuff that really tends to bring most great players down, which is never recovering from injuries and still having to have a huge role when you can't physically do it any more.

So, for posterity, I'll start with the per-100 stats comp b/w Kareem and Timmy, similar to the way I did the Shaq/Kareem comp. And, again, I'll emphasize that these box scores are NOT the crux of my argument...or even a primary component. They're there to give us some numbers to anchor the more important discussions to...really, just somewhere to start the conversation.

Kareem per-100 stats 1974: 29.9 PP100, 59.2 TS%, 3.1 ORBP100, 9.9DRP100; 4.5 AP100, 2.7 TO
22.0 FGA/100, 7.1 FTA/100 = ~25.1 possessions used on shots

Duncan per-100 stats career: 29.7 PP100, 55.1 TS%, 4.3 ORBP100, 12.6 DRP100; 4.7 AP100, 3.8 TO
22.8 FGA/100, 9.5 FTA/100 = ~27.0 possessions used on shots

In this comp, the overall volume per 100 is similar as scorers, and secondarily when factoring in possessions ending in assists and turnovers as well, but Duncan required a couple more shots and an extra turnover to do so. Maybe he got one of those possessions back with his advantage on the offensive boards, but still. And, as falcolombardi pointed out, Kareem's relative scoring efficiency for his time was even better than the raw TS% suggested. Fully stipulated.

But, as with the last comp, the real good stuff comes in when you start looking beyond the boxscores and into what's going on in the other 45 minutes (shoutout Bill Russell). I've never denied that Kareem was the better scorer than Duncan...if anything, I was a bit surprised that the per-100 stats were as close as they were. And, just like with Kareem and Shaq (but reversed), Duncan drew a steady diet of double teams and defensive attention in his career, but I'm fine with stipulating that Kareem was probably warping opposing defenses more than Duncan while also providing some good spacing with the range of his sky-hook.

But, unlike with the Shaq/Kareem comp, in this one we get to factor in what happens on the 100 defensive possessions as well.

Re: Duncan's defense

The discussion you posted about Duncan's defensive impact at the front and end of his career is interesting, and was worth fleshing out. And my response, is, two things can be true at the same time.

Yes, Duncan had an elite defensive supporting cast (and coaching staff) for the entirety of his career in San Antonio. I've been arguing that since I've been on this site, which (good grief) spans more than 2 decades now (old man in the house). It's why, for example, using ONLY team defensive rankings to argue Duncan's defensive excellence isn't the move.

But, the other truth is that Duncan was an elite defender in his own right as well. And even a quick look at those borders (e.g. the start to Duncan's career and the end) clearly shows Duncan's large impact. To whit, let's take a closer look at the borders mentioned in your post.

1996 (last Robinson year before Duncan): Spurs' team DRTG = 103.5 PP100, -4.1 from league average
1998 (first year Duncan): Spurs' team DRTG = 99.4, -5.6 relative to league average
bonus: 1999 (2nd Duncan): Spurs team DRTG 95.0, -7.2 relative to league average

2016 (last year Duncan): Spurs' team DRTG 99.0, -7.4 relative to league average
2017 (first year post-Duncan): Spurs team DRTG 103.5, -5.3 relative to league average

At both ends of his career, the Spurs were clearly a very strong defense even without Duncan. But with him, they were nigh historic. And improving an already elite unit to a clear degree is difficult to do. It helps demonstrate, even in a rough impact approach like this, that Duncan was having a huge defensive effect.

Similar approach to Kareem's offense:
I once did an impact comp of Kareem and Russell, utilizing a similar type of approach...trying to look at changes in unit efficacy at boundary points like this. I think the post was part of one of the previous Top 100 projects, but I also transferred it to my blog. If anyone wants to read the whole write-up, it's here: https://hoopslab.rotowire.com/post/162660433401/abdul-jabbar-vs-russell-observations-and-rough.

But, for the sake of this discussion, the numerical pattern wasn't necessarily Kareem's team offenses at the start/end of his career per se. Instead, I looked at Kareem's team offenses with and without all-history point guards. The thought, was, Russell's defensive impact showed up clearly in his team's defensive rating pattern from before his arrival, through his career, and then after, regardless of his teammates. But in Kareem's case, despite his awesome individual scoring/offense, the elite team offenses seemed to track much more with his all-history guard counterparts than with Kareem himself. To whit:

Bucks (2 years of Kareem, no Oscar): ORtg-rel: +1.7
Bucks (4 years of Kareem & Oscar): ORTg-rel: 4.2
Royals (Oscar’s entire career): ORtg-rel: +3.3

Lakers (4 years of Kareem, no Magic): ORtg-rel: +1.8
Lakers (10 years of Kareem & Magic): ORtg-rel: +4.9
Lakers (2 years of Magic, no Kareem): ORtg-rel: +5.1

Let me unpack this. In 6 years of prime/peak career when he wasn’t playing with Oscar or Magic, Kareem had monster boxscore numbers and multiple MVPs, but his team offenses were only +1.7 or +1.8. The offenses only got elite when the legendary point guards were around. And those offenses with Kareem + point guard played more similar to the offenses of those point guards without Kareem, than they did like Kareem’s offenses without them.

Way TL;DR. This all ties into/hopes to support the stance I've been arguing: Kareem was a historic scorer with strong defensive skills, but Duncan was a historic defensive player with strong scoring skills. In addition to the typical pattern, that it's better for a big man to be dominant on defense than offense, there is also some evidence that the pattern holds true for this specific case as well. What I haven't seen yet, and what I'd love someone to push back with, is much convincing evidence that this pattern DOESN'T hold true in this case.

[70sFan]

I'll start with some per-100 stats, even though that's not where my argument rests. But, it's good to have some numbers to reference in addition to the more description-based discussions below. So:

Kareem per-100 stats 1974: 29.9 PP100, 59.2 TS%, 3.1 ORBP100, 4.5 AP100, 2.7 TO
22.0 FGA/100, 7.1 FTA/100 = ~25.1 possessions used on shots

Shaq per-100 stats career: 35.2 PP100, 58.6 TS%, 5.2 ORBP100, 3.7 AP100, 4.1 TO
24.0 FGA/100, 13.9 FTA/100 = ~30.2 possessions used on shots

This sample doesn't include some of the most productive Kareem seasons though, which are 1970-73.

Worth noting that you don't include creation in used possessions. Shaq in his career has a negative AST/TOV ratio and even if we go with primes, 1998-03 Shaq had 1.13 ratio vs 1.23 for 1978-83 Kareem.

Kareem also played more minutes per game than Shaq and although Shaq scored more per possession, it's an era thing. No player in the 1970s posted as high usage as more modern players and I don't think it was caused by the limited abilites, but more so because of faster game and strategic differences.

Next, I want to post a well-known quote from Bill Russell:

If you take all the seconds added up shooting and rebounding, it comes to about three minutes. Now, out of a 48-minute game, three minutes are concerned with shooting and rebounding. What is going on the other 45 minutes?

The Russell quote shapes more of my thought process here than the numbers do, and hopefully give a frame of reference for how I see things. Because, if we assume roughly 100 possessions (not perfect, but round numbers), on both offense and defense, then what the box score numbers I posted tells us is that Kareem scored about 30 points on about 25 shooting possessions, while facilitating another 9ish points on 7ish passing possessions. Shaq generated about 5 more points on 5 more shooting possessions, and about 1.5 points less on 1 more passing possession.

Kareem's 39ish points on 32ish offensive possessions used, vs Shaq's 42.5ish points on 38ish possessions. I mean, we could debate which would be more valuable to a team. You could say Kareem was a bit more efficient on the whole, but Shaq was producing higher volume more directly as the scorer...but to me, the end result of these (average) 35ish possessions could get lost a bit in the noise, because, in the vein of Russell's quote...

Oh, now I see you counted creation, so I back off with my criticism

Kareem wasn't "a bit" more efficient, he was significantly more efficient once you adjust for eras differences:

Kareem: 114 TS+, 4719.0 TS Add, 3.0 TS Add/game
Shaq: 111 TS+, 2759.2 TS Add, 2.3 TS Add/game

The difference in primes is even more notable:

1971-80 Kareem: 116 TS+
1994-03 Shaq: 111 TS+

Of course, you can say that relative numbers don't tell the whole story, but the truth is that you couldn't produce nearly as good offense in 1970s without three point line and with lack of spacing. I think we should take that into account.

Another important thing is how player scores. Raw ppg or pts/poss don't tell us much, I am interested in how these points were scored. Here is a part of my Kareem vs Shaq post from peaks project, in which I go a bit deeper with scoring comparison:

I think even without adjusting for anything else, Kareem looks comfortably better to me, though raw volume difference does look significant. I want to touch a few points here:

1. Pace adjustments are very important in evaluations across eras, but we shouldn't stop at linear adjustments without taking into account the context behind these differences. Kareem's team played at much higher pace than Shaq, but we have to ask how much it actually helps Kareem's raw volume scoring numbers. Jabbar was a halfcourt player, who occasionally could score in transition. He's a post up center and to run your offense through him, you have to set your offense and start running plays. How much the increased number of transition possessions could help him? I'd say that Shaq played in an era that was the most suited to maximize low post scorers volume numbers - slow, very halfcourt-heavy offenses with few transition opportunities.

2. In postseason, Shaq averaged 30.6 pts/75 on +4.8 rTS% vs Kareem's 31.2 pts/75 on +13.7 rTS%. The difference in efficiency is staggering and it's not really related to small sample of size:

- Shaq's highest rTS% accomplished in the playoffs during his prime (1994-03) was +8.7 rTS%,
- Kareem surpassed that mark 6 times in 1970-83 period (1970, 1974, 1977, 1979, 1980, 1983).

I just think that Kareem could reach the level of efficiency (with similar volume) at levels that were beyond Shaq's reach. Efficiency also requires applying context, but in this case we're comparing two high volume post players who created the offense in similar way.

3. Shaq's scoring efficiency was heavily driven by putbacks and inside finishes. It could be seen both as advantage as disadvantage. On one hand, he's amazing at creating easy shots - better than Kareem. On the other, he's far more limited as a creator with the ball in his hands. Take a look at their post game numbers I tracked throughout the last year:

- 1971-79 Kareem (33 games): 21.8 ppg on 52.8 FG% and 57.1 TS%
- 2000-01 Shaq (38 games): 17.8 ppg on 49.3 FG% and 49.8 TS%

I think samples are decently representative for both. Again, it's up to you if you prefer Shaq's ability to generate easy points, or Kareem's ability to finish tough shots no matter what. I think what Kareem gives you brings a bit more value and is less teammates depended. We really haven't seen prime Shaq in a bad situation and I don't think he'd be able to carry his team to the same degree Kareem did. We also have seen Kareem in great situations (let's say in 1971 and 1980) and he showed ridiculous value, despite probably not being at his peak anymore.

Kareem was significantly better self-creator, it's not up to discussion.

What's going on in the other 65ish offensive possessions on offense? Or the 100 possessions on defense?

Since we're focusing only on the offensive end for this comp, let's forget about the 100 defensive possessions for now. On offense alone, we're still talking about 2/3 of the time that the bigs are on the court without finishing the possession.

One of the advancements basketball analysis that has fallen out of the longitudinal +/- studies is a series of descriptions of elements of the game that aren't measured in the box scores. Because, at least anecdotally, there doesn't seem to be a huge correlation between high volume/high efficiency finishing and the best offensive impacts. Instead, we've started using terms like "defense warping", "spacing" and (team) offense-creating to help explain what we tend to see. And, again anecdotally, these terms DO tend to correlate with the best offensive impacts.

Good point, we have to look way beyond boxscore numbers.

Bringing it back to Kareem vs Shaq
While Kareem was the most dominant big of his time, I think Shaq is the undisputed GOAT at warping opposing defenses. Opponents would play extra non-skilled bigs to use as fodder to try to slow him down (with 6 fouls, if nothing else), thus limiting their own offensive upside. They'd shade the entire team toward Shaq even when he didn't have the ball, then if he did get it he'd get routinely tripled in the post. Kenny Smith used to always say that Kobe Bryant was "the best 1-on-1 player in the world, that actually gets to play 1-on-1" because, as great as Kobe was, defenses had to overload on Shaq.

As much attention as Kareem drew, and he drew plenty, I think Shaq beats him handily in this aspect of the game. And by drawing so much defensive attention, he was creating better looks for his teammates simply by being on the court...to a higher degree than Kareem (or anyone). And he was doing this for all 65 of the possessions that he wasn't finishing with a shot, assist or TO.

I think here is where we disagree on that subject. When you watch 1974-79 Kareem, he drew so much defensive attention. His style was different, but I don't think defenses treated him any different than Shaq. Here is a part of my old post from peaks project:

About Shaq's gravity - this one is a massive game changer, but I wonder how much different it was compared to Kareem. I mean, this is how Kareem was guarded in 1977 playoffs:

Spoiler:

This are not highly selected screens - I picked them from one quarter of game 3 vs Warriors. Kareem absorbed ridiculous amount of defensive attention and he had a harder time beating it without the three point line.

I could be wrong, but Kareem drew so much defensive attention that I'm not willing to simply go with consensus, as most people don't really remember peak Kareem games. If you have any estimates proving me wrong, I will change my mind.

Similarly, I'd argue that Shaq's ability to warp a defense, multiple times, on every possession, over a full game, added up to a larger positive change in his team's offensive fortunes than any small differences one way or the other in the boxscore stats for the possessions that both actually used.

TL;DR: I argue that Shaq had more offensive impact than Kareem, not because he scored at a higher rate on similar efficiency, but because his overall offensive package (including and especially his ability to warp defenses) created more offensive opportunities for his team over the course of a game. And, therefore, Shaq's presence on the court would lead to a larger improvement in his team's offensive scoring margins than Kareem's would.

I think the main difference between them is the existance of three point line (I know, the line existed in the last seasons of Kareem's prime, but it was completely ignored by teams). I guess you can make a case that the existance of the three point line made Shaq doubles more impactful on team's offensive results. Kareem could be doubled or tripled, but all he could create is a long midrange shot which was low efficiency shot.

Now the question is - should we give Shaq credit for that in comparison to Kareem? Is there any reason to doubt that Kareem would create these open threes? I don't think so, but I could be wrong.

To sum up, I think that Kareem being significantly better shot creator, slightly better passer and overall clearly more efficient scorer is enough to put him ahead of Shaq's offensive rebounding advantage and foul drawing ability. I don't think it's clear enough to say that Shaq drew more attention than prime Kareem. From my experience, I wouldn't say that's the case but I didn't do hard studies on that.

Some people view Kareem as an all-time great finisher who has little value without the ball, but that's just a wrong picture. Kareem was very active off-ball and he pressured defenses as much as any post player could. On top of that, he was significantly more versatile offensive player than Shaq, who wasn't really an efficient post player in comparison and had very limited range.

[SinceGatlingWasARookie]

...

What makes you believe that Shaq was a better offensive player than Kareem?

I'll start with some per-100 stats, even though that's not where my argument rests. But, it's good to have some numbers to reference in addition to the more description-based discussions below. So:

Kareem per-100 stats 1974: 29.9 PP100, 59.2 TS%, 3.1 ORBP100, 4.5 AP100, 2.7 TO
22.0 FGA/100, 7.1 FTA/100 = ~25.1 possessions used on shots

Shaq per-100 stats career: 35.2 PP100, 58.6 TS%, 5.2 ORBP100, 3.7 AP100, 4.1 TO
24.0 FGA/100, 13.9 FTA/100 = ~30.2 possessions used on shots

Next, I want to post a well-known quote from Bill Russell:

If you take all the seconds added up shooting and rebounding, it comes to about three minutes. Now, out of a 48-minute game, three minutes are concerned with shooting and rebounding. What is going on the other 45 minutes?

The Russell quote shapes more of my thought process here than the numbers do, and hopefully give a frame of reference for how I see things. Because, if we assume roughly 100 possessions (not perfect, but round numbers), on both offense and defense, then what the box score numbers I posted tells us is that Kareem scored about 30 points on about 25 shooting possessions, while facilitating another 9ish points on 7ish passing possessions. Shaq generated about 5 more points on 5 more shooting possessions, and about 1.5 points less on 1 more passing possession.

Kareem's 39ish points on 32ish offensive possessions used, vs Shaq's 42.5ish points on 38ish possessions. I mean, we could debate which would be more valuable to a team. You could say Kareem was a bit more efficient on the whole, but Shaq was producing higher volume more directly as the scorer...but to me, the end result of these (average) 35ish possessions could get lost a bit in the noise, because, in the vein of Russell's quote...

What's going on in the other 65ish offensive possessions on offense? Or the 100 possessions on defense?

Since we're focusing only on the offensive end for this comp, let's forget about the 100 defensive possessions for now. On offense alone, we're still talking about 2/3 of the time that the bigs are on the court without finishing the possession.

One of the advancements basketball analysis that has fallen out of the longitudinal +/- studies is a series of descriptions of elements of the game that aren't measured in the box scores. Because, at least anecdotally, there doesn't seem to be a huge correlation between high volume/high efficiency finishing and the best offensive impacts. Instead, we've started using terms like "defense warping", "spacing" and (team) offense-creating to help explain what we tend to see. And, again anecdotally, these terms DO tend to correlate with the best offensive impacts.

Bringing it back to Kareem vs Shaq
While Kareem was the most dominant big of his time, I think Shaq is the undisputed GOAT at warping opposing defenses. Opponents would play extra non-skilled bigs to use as fodder to try to slow him down (with 6 fouls, if nothing else), thus limiting their own offensive upside. They'd shade the entire team toward Shaq even when he didn't have the ball, then if he did get it he'd get routinely tripled in the post. Kenny Smith used to always say that Kobe Bryant was "the best 1-on-1 player in the world, that actually gets to play 1-on-1" because, as great as Kobe was, defenses had to overload on Shaq.

As much attention as Kareem drew, and he drew plenty, I think Shaq beats him handily in this aspect of the game. And by drawing so much defensive attention, he was creating better looks for his teammates simply by being on the court...to a higher degree than Kareem (or anyone). And he was doing this for all 65 of the possessions that he wasn't finishing with a shot, assist or TO.

A few years back, Kirk Goldsberry presented a decision-making stat, where he used advanced scouting data to generate a decision-matrix for the ball-handler in every situation. It utilized a fractional approach, where adding up the fractional increases/decreases in a team's probability to score based on the decision-making of the ball-handler produced a quantified way to estimate how much such decision-making could add up over the course of a game.

Similarly, I'd argue that Shaq's ability to warp a defense, multiple times, on every possession, over a full game, added up to a larger positive change in his team's offensive fortunes than any small differences one way or the other in the boxscore stats for the possessions that both actually used.

TL;DR: I argue that Shaq had more offensive impact than Kareem, not because he scored at a higher rate on similar efficiency, but because his overall offensive package (including and especially his ability to warp defenses) created more offensive opportunities for his team over the course of a game. And, therefore, Shaq's presence on the court would lead to a larger improvement in his team's offensive scoring margins than Kareem's would.

Shaq warps a defense, but shot blockers to some degree warp an offense.

[Frosty]

So, for posterity, I'll start with the per-100 stats comp b/w Kareem and Timmy, similar to the way I did the Shaq/Kareem comp. And, again, I'll emphasize that these box scores are NOT the crux of my argument...or even a primary component. They're there to give us some numbers to anchor the more important discussions to...really, just somewhere to start the conversation.

Kareem per-100 stats 1974: 29.9 PP100, 59.2 TS%, 3.1 ORBP100, 9.9DRP100; 4.5 AP100, 2.7 TO
22.0 FGA/100, 7.1 FTA/100 = ~25.1 possessions used on shots

Duncan per-100 stats career: 29.7 PP100, 55.1 TS%, 4.3 ORBP100, 12.6 DRP100; 4.7 AP100, 3.8 TO
22.8 FGA/100, 9.5 FTA/100 = ~27.0 possessions used on shots

In this comp, the overall volume per 100 is similar as scorers, and secondarily when factoring in possessions ending in assists and turnovers as well, but Duncan required a couple more shots and an extra turnover to do so. Maybe he got one of those possessions back with his advantage on the offensive boards, but still. And, as falcolombardi pointed out, Kareem's relative scoring efficiency for his time was even better than the raw TS% suggested. Fully stipulated.

But, as with the last comp, the real good stuff comes in when you start looking beyond the boxscores and into what's going on in the other 45 minutes (shoutout Bill Russell). I've never denied that Kareem was the better scorer than Duncan...if anything, I was a bit surprised that the per-100 stats were as close as they were. And, just like with Kareem and Shaq (but reversed), Duncan drew a steady diet of double teams and defensive attention in his career, but I'm fine with stipulating that Kareem was probably warping opposing defenses more than Duncan while also providing some good spacing with the range of his sky-hook.

Your Per/100 stats miss the first 4 years of Kareem's career a period when he lead the league in scoring twice. They also factor in Kareem's last 2 years where he wasn't performing that well and at that age Duncan had already retired. This also factors in that Kareem was regularly putting in over 40 MPG during that period while Duncan never did and hovered around 34 mpg or less. You take a big and give him more time on the bench and he will have fresher legs. I'm sure Kareem's numbers would have been even better with more rest.

I'm kinda surprised you didn't consider those factors.

Personally I've never argued Kareem's offensive impact from a team perspective. Why? Because there were times where his coaches didn't build their offense around him. For instance the Showtime offense was a run and gun situation that often didn't even get into a half court set. It is hard on your center if he has to run back and forth while not getting touches. His value came in when they were forced into a half court set or needed a basket. He was lethal if they got him the ball and his FT shooting was good enough they wouldn't just put him on the line ala Shaq.

But, unlike with the Shaq/Kareem comp, in this one we get to factor in what happens on the 100 defensive possessions as well.

Re: Duncan's defense

The discussion you posted about Duncan's defensive impact at the front and end of his career is interesting, and was worth fleshing out. And my response, is, two things can be true at the same time.

Yes, Duncan had an elite defensive supporting cast (and coaching staff) for the entirety of his career in San Antonio. I've been arguing that since I've been on this site, which (good grief) spans more than 2 decades now (old man in the house). It's why, for example, using ONLY team defensive rankings to argue Duncan's defensive excellence isn't the move.

Agreed

But, the other truth is that Duncan was an elite defender in his own right as well. And even a quick look at those borders (e.g. the start to Duncan's career and the end) clearly shows Duncan's large impact. To whit, let's take a closer look at the borders mentioned in your post.

1996 (last Robinson year before Duncan): Spurs' team DRTG = 103.5 PP100, -4.1 from league average
1998 (first year Duncan): Spurs' team DRTG = 99.4, -5.6 relative to league average
bonus: 1999 (2nd Duncan): Spurs team DRTG 95.0, -7.2 relative to league average

2016 (last year Duncan): Spurs' team DRTG 99.0, -7.4 relative to league average
2017 (first year post-Duncan): Spurs team DRTG 103.5, -5.3 relative to league average

At both ends of his career, the Spurs were clearly a very strong defense even without Duncan. But with him, they were nigh historic. And improving an already elite unit to a clear degree is difficult to do. It helps demonstrate, even in a rough impact approach like this, that Duncan was having a huge defensive effect.

I'm unconvinced that Duncan's 1536 minutes his last year was the factor in them becoming 'nigh historic'.

Similar approach to Kareem's offense:

Similar but quite different.....

But, for the sake of this discussion, the numerical pattern wasn't necessarily Kareem's team offenses at the start/end of his career per se. Instead, I looked at Kareem's team offenses with and without all-history point guards. The thought, was, Russell's defensive impact showed up clearly in his team's defensive rating pattern from before his arrival, through his career, and then after, regardless of his teammates. But in Kareem's case, despite his awesome individual scoring/offense, the elite team offenses seemed to track much more with his all-history guard counterparts than with Kareem himself. To whit:

Bucks (2 years of Kareem, no Oscar): ORtg-rel: +1.7
Bucks (4 years of Kareem & Oscar): ORTg-rel: 4.2
Royals (Oscar’s entire career): ORtg-rel: +3.3

Lakers (4 years of Kareem, no Magic): ORtg-rel: +1.8
Lakers (10 years of Kareem & Magic): ORtg-rel: +4.9
Lakers (2 years of Magic, no Kareem): ORtg-rel: +5.1

Let me unpack this. In 6 years of prime/peak career when he wasn’t playing with Oscar or Magic, Kareem had monster boxscore numbers and multiple MVPs, but his team offenses were only +1.7 or +1.8. The offenses only got elite when the legendary point guards were around. And those offenses with Kareem + point guard played more similar to the offenses of those point guards without Kareem, than they did like Kareem’s offenses without them.

This actually lines up well with my earlier comparison of SAS defensive impacts with Robinson, Bowen and Leonard and how the Spurs dropped off without them.

Way TL;DR. This all ties into/hopes to support the stance I've been arguing: Kareem was a historic scorer with strong defensive skills, but Duncan was a historic defensive player with strong scoring skills. In addition to the typical pattern, that it's better for a big man to be dominant on defense than offense, there is also some evidence that the pattern holds true for this specific case as well. What I haven't seen yet, and what I'd love someone to push back with, is much convincing evidence that this pattern DOESN'T hold true in this case.

When Oscar joined Kareem he was on the downside of his career at 32 and only had a few years left. Kareem joining Mil catapulted them from the middle of the pack on offense (9th) to 2nd. When Kareem joined the Lakers they were 16 out of 18 offensively in the league and they immediately jumped to 7th.

I know the argument goes that Duncan was possibly the best defender ever but honestly he was never viewed that way in his prime and only after 'advanced stats' became a thing we see people bending over backwards to rewrite the narrative. He was a great defender but he played on great defensive teams. He also benefited from playing reduced minutes and not having to guard the best big on the opposing team. It's not like he played in an era where bigs weren't recognized for their defensive contributions lots were. These guys all won DPOY while Tim was playing

Mutombo
Mourning
Wallace
Camby
Garnett
Howard
Chandler
Gasol
Noah

You've been around awhile as have I. I find the recent Duncan discussions reminiscent of the old Kobe ones. You can try and have a well reasoned argument and there are people that just want to write you off as a hater or trying to 'tear down' someone. It's unfortunate that this goes on. I remember being firmly labelled as a Kobe hater for opinions that are pretty accepted today on the board.

[AEnigma]

People call you a hater because your lines of criticism are deeply disingenuous and often factually wrong, yet you repeat them anyway.

You can say Mutombo and Ben and peak Howard were better defenders than Duncan. Awards are not the end-all-be-all, but it is a fine comparison. However ignoring how voters dumbly and wrongly focused in on Bruce Bowen and how that myopia cost Duncan at least one DPoY (2007) and possibly more is not a coherent argument. Nor is pointing to how the Spurs defence declined without Bowen while completely disregarding how Bowen’s absence was also accompanied by Duncan’s own exit from his prime years. Robinson? The best Spurs defence happened without him. And as much as you vehemently insist that Duncan dodged any tough assessments, plenty of Spurs watchers have called you out for making that up (including myself).

[capfan33]

Re: Duncan's defense

The discussion you posted about Duncan's defensive impact at the front and end of his career is interesting, and was worth fleshing out. And my response, is, two things can be true at the same time.

Yes, Duncan had an elite defensive supporting cast (and coaching staff) for the entirety of his career in San Antonio. I've been arguing that since I've been on this site, which (good grief) spans more than 2 decades now (old man in the house). It's why, for example, using ONLY team defensive rankings to argue Duncan's defensive excellence isn't the move.

But, the other truth is that Duncan was an elite defender in his own right as well. And even a quick look at those borders (e.g. the start to Duncan's career and the end) clearly shows Duncan's large impact. To whit, let's take a closer look at the borders mentioned in your post.

1996 (last Robinson year before Duncan): Spurs' team DRTG = 103.5 PP100, -4.1 from league average
1998 (first year Duncan): Spurs' team DRTG = 99.4, -5.6 relative to league average
bonus: 1999 (2nd Duncan): Spurs team DRTG 95.0, -7.2 relative to league average

2016 (last year Duncan): Spurs' team DRTG 99.0, -7.4 relative to league average
2017 (first year post-Duncan): Spurs team DRTG 103.5, -5.3 relative to league average

At both ends of his career, the Spurs were clearly a very strong defense even without Duncan. But with him, they were nigh historic. And improving an already elite unit to a clear degree is difficult to do. It helps demonstrate, even in a rough impact approach like this, that Duncan was having a huge defensive effect.

Good summation of the effect Duncan had on defense, this is a great starting point for delving into this topic.

Similar approach to Kareem's offense:
I once did an impact comp of Kareem and Russell, utilizing a similar type of approach...trying to look at changes in unit efficacy at boundary points like this. I think the post was part of one of the previous Top 100 projects, but I also transferred it to my blog. If anyone wants to read the whole write-up, it's here: https://hoopslab.rotowire.com/post/162660433401/abdul-jabbar-vs-russell-observations-and-rough.

But, for the sake of this discussion, the numerical pattern wasn't necessarily Kareem's team offenses at the start/end of his career per se. Instead, I looked at Kareem's team offenses with and without all-history point guards. The thought, was, Russell's defensive impact showed up clearly in his team's defensive rating pattern from before his arrival, through his career, and then after, regardless of his teammates. But in Kareem's case, despite his awesome individual scoring/offense, the elite team offenses seemed to track much more with his all-history guard counterparts than with Kareem himself. To whit:

Bucks (2 years of Kareem, no Oscar): ORtg-rel: +1.7
Bucks (4 years of Kareem & Oscar): ORTg-rel: 4.2
Royals (Oscar’s entire career): ORtg-rel: +3.3

Lakers (4 years of Kareem, no Magic): ORtg-rel: +1.8
Lakers (10 years of Kareem & Magic): ORtg-rel: +4.9
Lakers (2 years of Magic, no Kareem): ORtg-rel: +5.1

Let me unpack this. In 6 years of prime/peak career when he wasn’t playing with Oscar or Magic, Kareem had monster boxscore numbers and multiple MVPs, but his team offenses were only +1.7 or +1.8. The offenses only got elite when the legendary point guards were around. And those offenses with Kareem + point guard played more similar to the offenses of those point guards without Kareem, than they did like Kareem’s offenses without them.

So this is where I think we have more disagreement. This is a similar approach but has a major difference because you're not actually comparing these teams before and after Kareem left them the way you did with Duncan. Doing so paints a different picture IMO. The year before Kareem came the Bucks played at about a -1 ORtg-rel (This is actually a bit conservative because they made a mid-season trade that improved them by about an 8 win pace, on bball reference they're at -1.5 for the whole season). A rookie Kareem (along with Dandridge) caused them to jump to a +3. I would say that's a pretty notable improvement that's at least comparable to Duncan, and this is with Jon McGlocklin as a PG lol. (Also their DRTG went from about a +2.3 to -1 as an aside).

In 75 the Bucks were 3-14 without Kareem but 35-30 with him, which is a pretty notable indicator of the type of lift Kareem had. They also lost Lucius Allen who had been one of their best players and his replacement only played 41 games. And according to Elgee's calculations, the Bucks were 7.5% PPA worse without Kareem than with Kareem. I'm not sure how that translates to ORTG, but I'm pretty confident that's a major difference.

The 76 Lakers were an awful team and an anomaly in this discussion. And the fit of the 78 and 79 teams was terrible because they were playing with a 3 guard starting lineup and one of the worst black holes in NBA history that lead to severe diminishing returns on offense.

And more generally, I think the overall point I made earlier still stands. I don't think it was possible for a big to have the type of impact Olajuwon and Shaq had on offense in the 90s due to spacing (as well as rules), without an excellent perimeter player next to them. The way the game was played back then simply didn't allow for it. Wilt, independent of the offensive issues he had, only led elite offenses on arguably the most loaded team ever up until that point. And Kobe was an elite perimeter player that was likely better than Oscar anyways by 2001 in regards to Shaq's offensive impact.

Overall I do think it's a philosophical difference to an extent, if you just take Kareem's numbers at face value they seem somewhat underwhelming. But while we can argue to what degree era mitigated Kareem's impact, I think it's undeniable that the offensive impact Kareem could have without a bonafide 2nd star next to him was capped in comparison to bigs in the 90s by virtue of not playing with a 3-point line and 5 out spacing. Especially in the case of Hakeem, not only was it era but he played with spacing that was well beyond that of most teams even at the same time period.

[PistolPeteJR]

And now Timmy fans are coming after Kareem. Or is it crypto Lebron fanscoming after Kareem?

Kareem was the better scorer, rebounder, arguably defender. He is the GOAT college player. He has a more extrnsive resume in general. And he was the the greater individual force on the basketball court.

You can always count on this guy to bring LeBron into it lol.

[capfan33]

And now Timmy fans are coming after Kareem. Or is it crypto Lebron fanscoming after Kareem?

Kareem was the better scorer, rebounder, arguably defender. He is the GOAT college player. He has a more extrnsive resume in general. And he was the the greater individual force on the basketball court.

You can always count on this guy to bring LeBron into it lol.

Duncan vs Kareem is a perfectly valid discussion lol.

[falcolombardi]

So, for posterity, I'll start with the per-100 stats comp b/w Kareem and Timmy, similar to the way I did the Shaq/Kareem comp. And, again, I'll emphasize that these box scores are NOT the crux of my argument...or even a primary component. They're there to give us some numbers to anchor the more important discussions to...really, just somewhere to start the conversation.

Kareem per-100 stats 1974: 29.9 PP100, 59.2 TS%, 3.1 ORBP100, 9.9DRP100; 4.5 AP100, 2.7 TO
22.0 FGA/100, 7.1 FTA/100 = ~25.1 possessions used on shots

Duncan per-100 stats career: 29.7 PP100, 55.1 TS%, 4.3 ORBP100, 12.6 DRP100; 4.7 AP100, 3.8 TO
22.8 FGA/100, 9.5 FTA/100 = ~27.0 possessions used on shots

In this comp, the overall volume per 100 is similar as scorers, and secondarily when factoring in possessions ending in assists and turnovers as well, but Duncan required a couple more shots and an extra turnover to do so. Maybe he got one of those possessions back with his advantage on the offensive boards, but still. And, as falcolombardi pointed out, Kareem's relative scoring efficiency for his time was even better than the raw TS% suggested. Fully stipulated.

But, as with the last comp, the real good stuff comes in when you start looking beyond the boxscores and into what's going on in the other 45 minutes (shoutout Bill Russell). I've never denied that Kareem was the better scorer than Duncan...if anything, I was a bit surprised that the per-100 stats were as close as they were. And, just like with Kareem and Shaq (but reversed), Duncan drew a steady diet of double teams and defensive attention in his career, but I'm fine with stipulating that Kareem was probably warping opposing defenses more than Duncan while also providing some good spacing with the range of his sky-hook.

Your Per/100 stats miss the first 4 years of Kareem's career a period when he lead the league in scoring twice. They also factor in Kareem's last 2 years where he wasn't performing that well and at that age Duncan had already retired. This also factors in that Kareem was regularly putting in over 40 MPG during that period while Duncan never did and hovered around 34 mpg or less. You take a big and give him more time on the bench and he will have fresher legs. I'm sure Kareem's numbers would have been even better with more rest.

I'm kinda surprised you didn't consider those factors.

Personally I've never argued Kareem's offensive impact from a team perspective. Why? Because there were times where his coaches didn't build their offense around him. For instance the Showtime offense was a run and gun situation that often didn't even get into a half court set. It is hard on your center if he has to run back and forth while not getting touches. His value came in when they were forced into a half court set or needed a basket. He was lethal if they got him the ball and his FT shooting was good enough they wouldn't just put him on the line ala Shaq.

But, unlike with the Shaq/Kareem comp, in this one we get to factor in what happens on the 100 defensive possessions as well.

Re: Duncan's defense

The discussion you posted about Duncan's defensive impact at the front and end of his career is interesting, and was worth fleshing out. And my response, is, two things can be true at the same time.

Yes, Duncan had an elite defensive supporting cast (and coaching staff) for the entirety of his career in San Antonio. I've been arguing that since I've been on this site, which (good grief) spans more than 2 decades now (old man in the house). It's why, for example, using ONLY team defensive rankings to argue Duncan's defensive excellence isn't the move.

Agreed

But, the other truth is that Duncan was an elite defender in his own right as well. And even a quick look at those borders (e.g. the start to Duncan's career and the end) clearly shows Duncan's large impact. To whit, let's take a closer look at the borders mentioned in your post.

1996 (last Robinson year before Duncan): Spurs' team DRTG = 103.5 PP100, -4.1 from league average
1998 (first year Duncan): Spurs' team DRTG = 99.4, -5.6 relative to league average
bonus: 1999 (2nd Duncan): Spurs team DRTG 95.0, -7.2 relative to league average

2016 (last year Duncan): Spurs' team DRTG 99.0, -7.4 relative to league average
2017 (first year post-Duncan): Spurs team DRTG 103.5, -5.3 relative to league average

At both ends of his career, the Spurs were clearly a very strong defense even without Duncan. But with him, they were nigh historic. And improving an already elite unit to a clear degree is difficult to do. It helps demonstrate, even in a rough impact approach like this, that Duncan was having a huge defensive effect.

I'm unconvinced that Duncan's 1536 minutes his last year was the factor in them becoming 'nigh historic'.

Similar approach to Kareem's offense:

Similar but quite different.....

But, for the sake of this discussion, the numerical pattern wasn't necessarily Kareem's team offenses at the start/end of his career per se. Instead, I looked at Kareem's team offenses with and without all-history point guards. The thought, was, Russell's defensive impact showed up clearly in his team's defensive rating pattern from before his arrival, through his career, and then after, regardless of his teammates. But in Kareem's case, despite his awesome individual scoring/offense, the elite team offenses seemed to track much more with his all-history guard counterparts than with Kareem himself. To whit:

Bucks (2 years of Kareem, no Oscar): ORtg-rel: +1.7
Bucks (4 years of Kareem & Oscar): ORTg-rel: 4.2
Royals (Oscar’s entire career): ORtg-rel: +3.3

Lakers (4 years of Kareem, no Magic): ORtg-rel: +1.8
Lakers (10 years of Kareem & Magic): ORtg-rel: +4.9
Lakers (2 years of Magic, no Kareem): ORtg-rel: +5.1

Let me unpack this. In 6 years of prime/peak career when he wasn’t playing with Oscar or Magic, Kareem had monster boxscore numbers and multiple MVPs, but his team offenses were only +1.7 or +1.8. The offenses only got elite when the legendary point guards were around. And those offenses with Kareem + point guard played more similar to the offenses of those point guards without Kareem, than they did like Kareem’s offenses without them.

This actually lines up well with my earlier comparison of SAS defensive impacts with Robinson, Bowen and Leonard and how the Spurs dropped off without them.

Way TL;DR. This all ties into/hopes to support the stance I've been arguing: Kareem was a historic scorer with strong defensive skills, but Duncan was a historic defensive player with strong scoring skills. In addition to the typical pattern, that it's better for a big man to be dominant on defense than offense, there is also some evidence that the pattern holds true for this specific case as well. What I haven't seen yet, and what I'd love someone to push back with, is much convincing evidence that this pattern DOESN'T hold true in this case.

When Oscar joined Kareem he was on the downside of his career at 32 and only had a few years left. Kareem joining Mil catapulted them from the middle of the pack on offense (9th) to 2nd. When Kareem joined the Lakers they were 16 out of 18 offensively in the league and they immediately jumped to 7th.

I know the argument goes that Duncan was possibly the best defender ever but honestly he was never viewed that way in his prime and only after 'advanced stats' became a thing we see people bending over backwards to rewrite the narrative. He was a great defender but he played on great defensive teams. He also benefited from playing reduced minutes and not having to guard the best big on the opposing team. It's not like he played in an era where bigs weren't recognized for their defensive contributions lots were. These guys all won DPOY while Tim was playing

Mutombo
Mourning
Wallace
Camby
Garnett
Howard
Chandler
Gasol
Noah

You've been around awhile as have I. I find the recent Duncan discussions reminiscent of the old Kobe ones. You can try and have a well reasoned argument and there are people that just want to write you off as a hater or trying to 'tear down' someone. It's unfortunate that this goes on. I remember being firmly labelled as a Kobe hater for opinions that are pretty accepted today on the board.

I wouldnt put too much weight in duncan not having a dpoy over thosr players

First because he probably deserved it at times over many of those guys (camby in particular)

Second because most of those guys won their dpoy's after duncan prime when he was in his mid 30's (noah, gasol, howard, chandler)

Third because garnett, mutombo and wallace are all time defenders themselves. And you could easily argue duncan to deserves a few of their dpoy's

And fourth and most importantly because having or not a dpoy shouldnt bear much weight by itself. Is an award that alvin robertson has but tim duncan doesnt

[AEnigma]

I wouldnt put too much weight in duncan not having a dpoy over thosr players

First because he probably deserved it at times over many of those guys (camby in particular)

[…]

And fourth and most importantly because having or not a dpoy shouldnt bear much weight by itself. Is an award that alvin robertson has but tim duncan doesnt

I do not know, how could Hakeem be the best defender since Russell when he could not even beat out Alvin, Cooper, or Rodman?

[falcolombardi]

I'll start with some per-100 stats, even though that's not where my argument rests. But, it's good to have some numbers to reference in addition to the more description-based discussions below. So:

Kareem per-100 stats 1974: 29.9 PP100, 59.2 TS%, 3.1 ORBP100, 4.5 AP100, 2.7 TO
22.0 FGA/100, 7.1 FTA/100 = ~25.1 possessions used on shots

Shaq per-100 stats career: 35.2 PP100, 58.6 TS%, 5.2 ORBP100, 3.7 AP100, 4.1 TO
24.0 FGA/100, 13.9 FTA/100 = ~30.2 possessions used on shots

This sample doesn't include some of the most productive Kareem seasons though, which are 1970-73.

Worth noting that you don't include creation in used possessions. Shaq in his career has a negative AST/TOV ratio and even if we go with primes, 1998-03 Shaq had 1.13 ratio vs 1.23 for 1978-83 Kareem.

Kareem also played more minutes per game than Shaq and although Shaq scored more per possession, it's an era thing. No player in the 1970s posted as high usage as more modern players and I don't think it was caused by the limited abilites, but more so because of faster game and strategic differences.

Next, I want to post a well-known quote from Bill Russell:

If you take all the seconds added up shooting and rebounding, it comes to about three minutes. Now, out of a 48-minute game, three minutes are concerned with shooting and rebounding. What is going on the other 45 minutes?

The Russell quote shapes more of my thought process here than the numbers do, and hopefully give a frame of reference for how I see things. Because, if we assume roughly 100 possessions (not perfect, but round numbers), on both offense and defense, then what the box score numbers I posted tells us is that Kareem scored about 30 points on about 25 shooting possessions, while facilitating another 9ish points on 7ish passing possessions. Shaq generated about 5 more points on 5 more shooting possessions, and about 1.5 points less on 1 more passing possession.

Kareem's 39ish points on 32ish offensive possessions used, vs Shaq's 42.5ish points on 38ish possessions. I mean, we could debate which would be more valuable to a team. You could say Kareem was a bit more efficient on the whole, but Shaq was producing higher volume more directly as the scorer...but to me, the end result of these (average) 35ish possessions could get lost a bit in the noise, because, in the vein of Russell's quote...

Oh, now I see you counted creation, so I back off with my criticism

Kareem wasn't "a bit" more efficient, he was significantly more efficient once you adjust for eras differences:

Kareem: 114 TS+, 4719.0 TS Add, 3.0 TS Add/game
Shaq: 111 TS+, 2759.2 TS Add, 2.3 TS Add/game

The difference in primes is even more notable:

1971-80 Kareem: 116 TS+
1994-03 Shaq: 111 TS+

Of course, you can say that relative numbers don't tell the whole story, but the truth is that you couldn't produce nearly as good offense in 1970s without three point line and with lack of spacing. I think we should take that into account.

Another important thing is how player scores. Raw ppg or pts/poss don't tell us much, I am interested in how these points were scored. Here is a part of my Kareem vs Shaq post from peaks project, in which I go a bit deeper with scoring comparison:

I think even without adjusting for anything else, Kareem looks comfortably better to me, though raw volume difference does look significant. I want to touch a few points here:

1. Pace adjustments are very important in evaluations across eras, but we shouldn't stop at linear adjustments without taking into account the context behind these differences. Kareem's team played at much higher pace than Shaq, but we have to ask how much it actually helps Kareem's raw volume scoring numbers. Jabbar was a halfcourt player, who occasionally could score in transition. He's a post up center and to run your offense through him, you have to set your offense and start running plays. How much the increased number of transition possessions could help him? I'd say that Shaq played in an era that was the most suited to maximize low post scorers volume numbers - slow, very halfcourt-heavy offenses with few transition opportunities.

2. In postseason, Shaq averaged 30.6 pts/75 on +4.8 rTS% vs Kareem's 31.2 pts/75 on +13.7 rTS%. The difference in efficiency is staggering and it's not really related to small sample of size:

- Shaq's highest rTS% accomplished in the playoffs during his prime (1994-03) was +8.7 rTS%,
- Kareem surpassed that mark 6 times in 1970-83 period (1970, 1974, 1977, 1979, 1980, 1983).

I just think that Kareem could reach the level of efficiency (with similar volume) at levels that were beyond Shaq's reach. Efficiency also requires applying context, but in this case we're comparing two high volume post players who created the offense in similar way.

3. Shaq's scoring efficiency was heavily driven by putbacks and inside finishes. It could be seen both as advantage as disadvantage. On one hand, he's amazing at creating easy shots - better than Kareem. On the other, he's far more limited as a creator with the ball in his hands. Take a look at their post game numbers I tracked throughout the last year:

- 1971-79 Kareem (33 games): 21.8 ppg on 52.8 FG% and 57.1 TS%
- 2000-01 Shaq (38 games): 17.8 ppg on 49.3 FG% and 49.8 TS%

I think samples are decently representative for both. Again, it's up to you if you prefer Shaq's ability to generate easy points, or Kareem's ability to finish tough shots no matter what. I think what Kareem gives you brings a bit more value and is less teammates depended. We really haven't seen prime Shaq in a bad situation and I don't think he'd be able to carry his team to the same degree Kareem did. We also have seen Kareem in great situations (let's say in 1971 and 1980) and he showed ridiculous value, despite probably not being at his peak anymore.

Kareem was significantly better self-creator, it's not up to discussion.

What's going on in the other 65ish offensive possessions on offense? Or the 100 possessions on defense?

Since we're focusing only on the offensive end for this comp, let's forget about the 100 defensive possessions for now. On offense alone, we're still talking about 2/3 of the time that the bigs are on the court without finishing the possession.

One of the advancements basketball analysis that has fallen out of the longitudinal +/- studies is a series of descriptions of elements of the game that aren't measured in the box scores. Because, at least anecdotally, there doesn't seem to be a huge correlation between high volume/high efficiency finishing and the best offensive impacts. Instead, we've started using terms like "defense warping", "spacing" and (team) offense-creating to help explain what we tend to see. And, again anecdotally, these terms DO tend to correlate with the best offensive impacts.

Good point, we have to look way beyond boxscore numbers.

Bringing it back to Kareem vs Shaq
While Kareem was the most dominant big of his time, I think Shaq is the undisputed GOAT at warping opposing defenses. Opponents would play extra non-skilled bigs to use as fodder to try to slow him down (with 6 fouls, if nothing else), thus limiting their own offensive upside. They'd shade the entire team toward Shaq even when he didn't have the ball, then if he did get it he'd get routinely tripled in the post. Kenny Smith used to always say that Kobe Bryant was "the best 1-on-1 player in the world, that actually gets to play 1-on-1" because, as great as Kobe was, defenses had to overload on Shaq.

As much attention as Kareem drew, and he drew plenty, I think Shaq beats him handily in this aspect of the game. And by drawing so much defensive attention, he was creating better looks for his teammates simply by being on the court...to a higher degree than Kareem (or anyone). And he was doing this for all 65 of the possessions that he wasn't finishing with a shot, assist or TO.

I think here is where we disagree on that subject. When you watch 1974-79 Kareem, he drew so much defensive attention. His style was different, but I don't think defenses treated him any different than Shaq. Here is a part of my old post from peaks project:

About Shaq's gravity - this one is a massive game changer, but I wonder how much different it was compared to Kareem. I mean, this is how Kareem was guarded in 1977 playoffs:

Spoiler:

This are not highly selected screens - I picked them from one quarter of game 3 vs Warriors. Kareem absorbed ridiculous amount of defensive attention and he had a harder time beating it without the three point line.

I could be wrong, but Kareem drew so much defensive attention that I'm not willing to simply go with consensus, as most people don't really remember peak Kareem games. If you have any estimates proving me wrong, I will change my mind.

Similarly, I'd argue that Shaq's ability to warp a defense, multiple times, on every possession, over a full game, added up to a larger positive change in his team's offensive fortunes than any small differences one way or the other in the boxscore stats for the possessions that both actually used.

TL;DR: I argue that Shaq had more offensive impact than Kareem, not because he scored at a higher rate on similar efficiency, but because his overall offensive package (including and especially his ability to warp defenses) created more offensive opportunities for his team over the course of a game. And, therefore, Shaq's presence on the court would lead to a larger improvement in his team's offensive scoring margins than Kareem's would.

I think the main difference between them is the existance of three point line (I know, the line existed in the last seasons of Kareem's prime, but it was completely ignored by teams). I guess you can make a case that the existance of the three point line made Shaq doubles more impactful on team's offensive results. Kareem could be doubled or tripled, but all he could create is a long midrange shot which was low efficiency shot.

Now the question is - should we give Shaq credit for that in comparison to Kareem? Is there any reason to doubt that Kareem would create these open threes? I don't think so, but I could be wrong.

To sum up, I think that Kareem being significantly better shot creator, slightly better passer and overall clearly more efficient scorer is enough to put him ahead of Shaq's offensive rebounding advantage and foul drawing ability. I don't think it's clear enough to say that Shaq drew more attention than prime Kareem. From my experience, I wouldn't say that's the case but I didn't do hard studies on that.

Some people view Kareem as an all-time great finisher who has little value without the ball, but that's just a wrong picture. Kareem was very active off-ball and he pressured defenses as much as any post player could. On top of that, he was significantly more versatile offensive player than Shaq, who wasn't really an efficient post player in comparison and had very limited range.

Only thingh i would add to your comment: illegal defense

Arguably even more important that the spacing advantage (not that the extra bit of spacing from 3 point shooters didnt help shaq) is that teans had more limited ways to defend him due to ruleset

If kareem played with illegal D rules in the 70's it would have made getting him the ball easier and blitzing/doubling him effectively tougher. And if he had 2-3 3 point shooters on top of that? Even harder

One of the "simple" thinghs that got me higher on kareem that i expectes was watching him vs walton in 77. Not because he made walton post defense look like a prop (although was bad ass too)

But because walton would -leave the paint- just to prevent kareem from catching. I cannot understate this

In a era of no 3 pointers and bad spacing when the paint was the utmost priority to a defense (even more than today) and even the likes of bill russel who had the mobility to play deep in the perimeter prefered to concede long midrange shots of marginal value most of the time rather than leaving the paint area unprotected

Walton, one of the greatest and smartest rim protectors to ever play would leave it completely unoccupied to faceguard/fronting kareem. He was more worried about kareem post ups that a unguarded rim to attack

This is curry level off ball gravity impact

[Texas Chuck]

Yeah whining about people calling you a hater when you express the following seems odd:

Bowen a more important defender than Duncan
Duncan never guards the best opposing big
Duncan never won a DPOY so how good was he really?

I don't think you honestly believe the first two are true and I know you realize the 3rd is of zero relevance to anything unless you think Shaq was only the most valuable player in the league 1 season when we all know better than that.

It's fine to be lower on Duncan than most. Nobody has any issues with that. Place would suck if we all hold the same opinions. But poor arguments will always be pushed back on. And when the same tired disproved arguments are used time and again, people are going to start to wonder the motivation for it.

[Cavsfansince84]

And now Timmy fans are coming after Kareem. Or is it crypto Lebron fanscoming after Kareem?

Kareem was the better scorer, rebounder, arguably defender. He is the GOAT college player. He has a more extrnsive resume in general. And he was the the greater individual force on the basketball court.

You can always count on this guy to bring LeBron into it lol.

Duncan vs Kareem is a perfectly valid discussion lol.

I think it is but at the same time I feel its valid to say that Kareem's prime was half a tier above Duncan's. I think Tim really needed to continue at his 01-early 05 pace for a few more years to truly be on Kareem's level who did that from 71-80.

[Stalwart]

And now Timmy fans are coming after Kareem. Or is it crypto Lebron fanscoming after Kareem?

Kareem was the better scorer, rebounder, arguably defender. He is the GOAT college player. He has a more extrnsive resume in general. And he was the the greater individual force on the basketball court.

You can always count on this guy to bring LeBron into it lol.

Duncan vs Kareem is a perfectly valid discussion lol.

Kareem was transcendent. Duncan was not. Two different tiers of players.

[falcolombardi]

You can always count on this guy to bring LeBron into it lol.

Duncan vs Kareem is a perfectly valid discussion lol.

Kareem was transcendent. Duncan was not. Two different tiers of players.

Many posters would disagree

What would be your reasons to think kareem is in a different tier to duncan?

[SNPA]

I didn’t go through this whole thread so apologies if this has been stated.

The answer is, play in the 70’s.

Also, I have them fairly close anyways.

[70sFan]

You can always count on this guy to bring LeBron into it lol.

Duncan vs Kareem is a perfectly valid discussion lol.

Kareem was transcendent. Duncan was not. Two different tiers of players.

I can understand all these takes about Hakeem, but now Duncan? Seriously? Duncan has better resume than Kobe freaking Bryant, that's enough to make him trascendent.