Re: Kareem's scoring dominance vs modern bigs (box score volume comp; argument for pace-adjusting)Below, I'll paste a recent post from a thread that was meant to compare Kareem with Duncan and Garnett on the basis of longevity and dominance. Before I do, though, I want to speak about pace adjusting. I think, if boxscore volume in any way influences your analysis, some version of pace adjusting is absolutely VITAL. But, I had a conversation with Blackmill in response to the post below, that made me realize that I should clarify a bit WHY I believe pace adjusting is so important.
I've seen others post in recent years similarly to Blackmill did, essentially arguing that pace adjusting isn't an accurate means of projecting exactly what one player would produce in different circumstances. That, since prominent/iso scorers generally get much of their production in the half-court, their production doesn't ramp up linearly with pace. And I buy that, to some extent.
HOWEVER
I also think that pace HAS to be accounted for, as far as volume stats go, just for common sense sake. If one player gets 120 possessions a game and another gets 90, there's NO WAY you can just judge their relative volume stats as though they are on the same plane. One guy got 33% more opportunity than the other...you can't just ignore that, if the raw volume numbers mean anything to you.
So, if both stances have merit, what't the solution? Well, what I realized when debating with Blackmill, was, I don't actually CARE that pace-adjusting doesn't necessarily accurately predict a player's volume if he switched circumstances. Because if they switch circumstances, there's a LOT of variables that change. Swapping Duncan with Kareem, for example, completely changes the types of offensive schemes that are run, the types of defensive schemes that are run, the pace, the 3-point line, other rule changes, the way that rules are enforced, the types of training techniques available, the economic realities of professional athletes then vs now, etc. etc. etc. There are WAY too many variables to accurately predict, with numbers, what Kareem's stats would have looked like for the 2003 Spurs or what Duncan's stats would have looked like for the 1977 Lakers.
But...THAT'S NOT THE POINT!
I don't want to know exactly what each would produce in the other's shoes, there's no way to know that. Instead, I want to use their volumes, in their given eras, to estimate how much weight each was carrying in their own time! And to THAT end, I think pace adjusting is vital. Because while 1977 Kareem had a higher scoring average than 2003 Duncan by a few points per game, it actually turns out that
1977 Kareem scored 24.5% of the Lakers' points that year while Duncan in 2003 scored 24.3% of the Spurs' points. That is virtually identical, and a much more accurate representation of how much scoring load each was carrying for their team than the raw totals. And, bringing it full circle, that type of equality in scoring load carried is revealed in pace-adjusted numbers.
Kareem is getting a lot of votes in this thread, whereas I haven't seen a Duncan vote yet. And while I could be wrong, I strongly believe that a) the incumbent factor (e.g. we've believed in Kareem's inherent greatness for a LONG time) and b) the raw numbers that suggest Kareem's scoring dominance are very, very influential in the votes so far. So...let me repost what I put in the other thread. Maybe it doesn't influence anyone here at all, but for me it really forces me to think. WAS Kareem really more dominant than modern bigs like Duncan? Or are his numbers just better because the era in which he played was more conducive to big volume numbers (this same argument will relate to Wilt, Oscar, West, and everyone else from that era)
Without further ado, here's a re-post from this location:
viewtopic.php?p=56233070#p56233070 Re: Kareem vs Duncan (and a bit of Garnett)There are some interesting things, here. I did some very basic looks at stats today, and I think there are some definite things that are accepted as givens that should be re-thought.
1) Longevity. Kareem had amazing longevity, but upon examination there's a reasonable argument that Duncan's is better. Consider (purely boxscore measurements, per 100 possessions where available):
Year 11970 Kareem (not per 100): 28.8 pts (55.2% TS), 14.5 reb, 4.1 ast --> 22.5 PER, .187 WS/48
1998 Duncan (per 100 pos): 29.3 pts (57.5% TS), 16.6 reb, 3.8 ast--> 22.6 PER, .192 WS/48
Kareem played 82 games, 43 min/game. Duncan played 82 games, 39 minutes.
Year 181987 Kareem (per 100): 26.4 pts (59.7% TS), 10.1 reb, 3.9 ast, 3.6 TO -> 1.2 OBPM, 0.4 DBPM
2015 Duncan (per 100): 24.6 pts (56.0% TS), 16.2 reb, 5.3 ast, 3.0 TO -> 0.8 OBPM, 4.7 DBPM
Kareem played 78 games, 31 minutes a game. Duncan played 77 games, 29 minutes a game.
Now, you guys all know that the boxscore alone isn't nearly enough for me to feel good about an evaluation. Also, per 100 numbers aren't perfect for analyzing different eras, but it does provide some type of pace normalization which I think is necessary. But with those said...
Are we sure that Kareem's longevity is better? Duncan's year 18 certainly looks more impressive in the regular season boxscores, to me. Slight advantage to Kareem as a scorer, but Duncan was a SIGNIFICANTLY better rebounder and also a much more efficient distributor at this point in their careers. And, stepping outside of the boxscores, we know that Duncan was still an elite defender at that point in his career and Kareem wasn't.
At the moment, I'm seeing longevity as, at worst, a wash and at best a slight advantage for DUNCAN when compared to Kareem.
2) Peak play.Again, just a quick boxscore analysis from regular season, but with pace adjusted number. Also, per the OP, let's throw KG in there:
1977 Kareem: 32.7 points/100 (60.8% TS), 16.6 reb, 4.8 ast --> 7.7 OBPM, 3 DBPM
2003 Duncan: 31.6 points/100 (56.4% TS), 17.5 reb, 5.3 ast--> 3.3 OBPM, 4.2 DBPM
2004 Garnett: 33.2 points/100 (54.7% TS), 19.0 reb, 6.8 ast -->4.9 OBPM, 5 DBPM
Same disclaimers as before. Boxscore isn't enough, we're looking at regular season only, pace adjusting isn't perfect...all of that.
That said again...perfect or not, the pace adjusting really puts things in a different perspective. Because much of Kareem's claim to "more dominance" than Duncan or KG comes from his mega scoring. But he was also playing at the breakneck pace of the 1970s vs the grind-it-out early 2000s and...perfect or not, the per 100 stats indicate clearly that at their peaks Duncan and Garnett were scoring at volumes (relative to their teammates & pace) that were very similar to where Kareem was operating.
Way more would need to go into a real analysis of who was better at their peaks than just this. But with this as a reasonable first cut look...I see absolutely no reason why it wouldn't be reasonable to find after looking deeper that Duncan or Garnett (or both) may have actually peaked higher than Kareem.
Conclusion: I really hope that this type of conversation does help people to look into this/these comps in more depth and not just go with the default. Because there is some legitimate "there" there. Longevity, dominance and scoring are Kareem's calling cards...but they aren't givens, here.
