RealGM 2023 Top 100 Project - #5 (Tim Duncan)

[One_and_Done]

Vote

1. Tim Duncan
2. Shaq

Same vote as last time - excellent longevity, consistent career, one of the best leadership impacts of all time, and good enough peak with runs like 03 playoffs.

Wilt was shakiest in playoffs compared to regular season of these players I believe, and was a difficult personality behind the scenes. I like pre-93 Hakeem less than the secondary Duncan years and he has less longevity. Magic is a slight era downgrade for me and less longevity. Shaq is hardest since he has strengths and weaknesses compared to Duncan but I think the latter's value is more consistent.

Nominate: Stephen Curry

I'm probably going to have KG top 10, but I think it's more important to get Curry in the voter pool first. KG, KD and Bird will be the next 3 names for nomination for me after Curry. After that it'll get tough.

[eminence]

Looking like a close race in the nomination bracket between Bird, KG and Curry.

[iggymcfrack]

Vote: Hakeem Olajuwon

This is a big shift for me and easily the most I've changed my mind in one discussion period. I always want people to be open to my arguments though and so I feel it's only fair to be open to changing my mind when presented with compelling evidence. After being initially impressed with f4p's post, there were some things that I investigated myself:

-First, Hakeem did have significantly better stats than Duncan in the playoffs throughout his career. His edge in the postseason was similar to Duncan's edge in the regular season. Hakeem had a career BPM of 6.9 in the playoffs compared to 5.9 for Duncan and this advantage holds even if you use an age cutoff to keep Duncan from having so many playoff games past his prime, late in his career.

-Both WOWY and WOWYR data give Hakeem the edge and there's a significant sample for both. For instance, when Hakeem missed the playoffs in 1992, the Rockets went 40-30 with him and 2-10 without him. This is an even wider split than Kareem had in 74/75 when the Bucks missed the playoffs and is very understandable. Hakeem didn't miss enough games for it to be a problem, but he missed enough to see that yes, most of the time, he did have a very poor supporting cast in Houston and I can't really blame him for failing to elevate his team consistently to the levels that Duncan did.

-I also feel relatively confident that Hakeem had a significant defensive edge. He didn't play on as good of defensive teams as Duncan and didn't always put up eye-popping rDRtg numbers on a team-wide basis, but I do feel that I would have to grade him as a meaningfully better defender overall based on multiple educated opinions as well as the few data points that we do have.

-I don't feel that Duncan had a meaningful edge in longevity. Duncan played 19 seasons to Hakeem's 18 and who had more meaningfully impactful seasons kinda depends on where the cutoff is for meaningful. If you just take their top 10 seasons away, Hakeem might gain more in seasons 11-13 than he loses in seasons 16-19 is 14 and 15 are roughly equal.

So when it comes down to it, I'd say:
Scoring: Hakeem > Duncan
Passing: Duncan > Hakeem
Defense: Hakeem > Duncan

And there's no way the difference in passing is enough to overcome the other two. Both did an absolutely A+ job of being clutch winners within the constraints they played in and both were a near push on longevity. I think I have to give Hakeem the edge as the superiorly skilled player by the slightest of margins.

Alternate: Tim Duncan


Nominate: Kevin Garnett: Tremendous player, analytics darling, and honestly, the one other player I could possibly be convinced to put ahead of Tim Duncan in this project. Somehow manages to show out as somewhere between the 2nd and 4th most valuable player of the past 25 years in pretty much every dataset despite being used completely wrong as a defender for much of his Minnesota years, taking away from an all-time strength as a help defender to position him on the perimeter.

[ijspeelman]

Vote: Tim Duncan



I do not have time to make a long film post this week so I am mainly just going to quote previous things I said about Duncan as my explanation.

In the early parts of his career, Duncan combined dominant post scoring with elite post and help defense. This era mainly consisted of the illegal defense rules so both post scoring and post defense were at a premium. To me this all culminates with the 2002-03 Spurs champion where he led a team with an aging Robinson, a young Parker, and a young Ginobli.

After the illegal defense rules were abolished, offense became more and more perimeter based, but even so most teams did not abandon the post. Duncan kept up his post scoring and efficiency for almost his entire career, albeit at a slightly lesser rate than his two MVP seasons. To touch on it quickly, Duncan’s spacing with his midrange is often overstated. That 10-16 footer that he shot as 20% of his FGs only went in 40% of the time.

Duncan will always be brought up with his stellar teammates, the aforementioned Robinson, Parker, and Ginobli, but more importantly with Popovich. This connection seems to downplay Duncan unfairly. He was graced with great teams which then resulted in great team success, yes, but without his dominant defense and scoring, the Spurs would not be regarded as they are today. You can look at prime David Robinson for what it looks like when a star Spur doesn’t have a great team.

What I've noted though over the years in this project is that Duncan didn't gain separation from Hakeem & Shaq until the 2014 project. Maybe if we'd done a 2013 project he'd have gotten the separation there, but he wasn't in 2011 and I'm pretty confident '11-12 wouldn't have changed this.

The gaining of separation based on the Beautiful Game years of the Spurs doesn't necessarily strike me as problematic...but when I start seeing some of the things I've seen written, well, I have concerns.

I think what separates someone like Duncan from someone like Shaq is exactly because he became a productive all-star to sub all-star later in his career. Duncan transformed his dominant MVP seasons into perennial all defensive seasons while providing the connectivity that those later Spurs teams needed. Around the 2010-12 mark, Duncan did decline in his impact on the team, but his willingness to let that happen and just be hyper focused on his role still made him a solid all-star level big man.

Whereas with someone like Shaq, even prior to his Heat championship he was faltering quickly. Now, he was still near an all-star talent in his Heat days, but soon his offense was not worth his defense and his offense was also not nearly as productive. I give Duncan a lot of credit for being a near all-star to an MVP for the entirety of his career.

To me, [Bill Russell and Tim Duncan] are versions of each other, but in different eras. They both have their relative longevity with prolonged team success. Where they differ is that Timmy was a complete offensive threat. In his first six seasons (where I have him as the all out offensive hub for the Spurs), his teams offense, while not incredible, were either middle of the pack or for 3 seasons in the top third (ranking as high as 6, 7, and 9 out of the 29 teams). Whereas Russell was not relied on for offense and for good reason. Russell was near-ish league average TS% on much below league after FGA/36 minutes. His team's also ranked below league average on offense for his entire career besides one year (they also ranked dead last 3/13 times).

But, his defense makes up for it. Due to his era (no three point line or illegal defense), the center position was the most important position defensively (its not so different today, but it was magnitudes more important in Russell's day) and Russell was the best defender. He disrupted, he blocked, he deterred, and most importantly he ran up and down the floor and across the key constantly. This made him, in my opinion, the greatest defensive player ever. Now, Duncan is no snub himself and probably belongs as one of, if not the best, defensive player of the 2000s. Duncan was quick and athletic with a huge body and arms. He had a brain like Russell and was constantly in the right spot at the right time.

Nomination: Kevin Garnett

Spoiler:

A lot of these lower spots are dominated by big men and rightfully so. During most of NBA history, if you were an exceptional big man defender and brought anything on the offensive side you were a consistent all-star to all-NBA talent. Garnett brings modern-like switchability, help-side defense, and constant disruption at the big man spot. He also brought what most people credit Duncan with, spacing from the big man in an era with typically poor shooting. Where Duncan trumps him is in the post, both offensively and defensively. However, with Garnett's skillset I think he deserves a spot in our 5 contenders for the #4 spot.

I honestly think that Garnett is going to be a near vote for me as soon as he's nominated.

[OldSchoolNoBull]

Longevity

I want to first address the longevity issue. Seeing as they both played nineteen years, I think Shaq's longevity hit(i.e. the extent to which he declined towards the end) is, while not inaccurate, overstated, while I also believe there's one key deficiency in Duncan's longevity that is being overlooked.

Look at their respective season-by-season RS rTS numbers.

Shaq:

Code: Select all

93 / +4.9
94 / +7.7
95 / +4.5
96 / +2.8
97 / +2.0
98 / +6.3
99 / +7.3
00 / +5.5
01 / +5.6
02 / +7.0
03 / +8.3
04 / +6.2
05 / +5.4
06 / +5.0
07 / +2.6
08 / +4.9
09 / +7.9
10 / +2.2
11 / +11.8


Duncan:

Code: Select all

98 / +5.2
99 / +3
00 / +3.2
01 / +1.8
02 / +5.4
03 / +4.5
04 / +1.8
05 / +1.1
06 / -1.3
07 / +3.8
08 / +0.6
09 / +0.5
10 / +1.7
11 / -0.4
12 / +0.4
13 / +1.9
14 / -0.6
15 / +2.6
16 / -1.8


I made this point in a previous post, but Duncan's scoring efficiency fell off significantly after his first six years. He was a legitimately top-tier scorer in the first part of his career, but by the mid-00s he wasn't the same scorer. As you can see above, from 2004 onward, he topped +2 rTS only twice, had a <1 rTS seven times, and a negative rTS four times.

His career average TS Add is 53.1. His average TS add during his first six years is 122.5. His average TS Add during the the last thirteen years of his career is 21.3. If we adjust the ranges a bit and look at his first ten years(up through 2006-07) and his last nine years - he has an average TS Add of 87.9 over the first ten years and 14.5 over the last nine.

I realize that some of the decrease in TS Add can be attributed to a decrease in volume as a result of a decrease in minutes, but that doesn't explain the efficiency drops as illustrated above.

Duncan's defense and playmaking ability were effective until the very end, but his longevity as a scorer seems not great.

On Shaq's end, there's no question he declined a lot defensively, and that he was slow and not moving very well for his last few seasons, but his bread and butter was always scoring, and in that respect I'm not sure he was ever ineffective. He kept up healthy rTS numbers right up until the end, and as late as 2008-09, he was putting up 17 points(and 8 boards) on +7.9% rTS over 75 games in Phoenix at the age of 36. You can chalk that up to Steve Nash magic if you'd like, but those are still solid(in volume)/elite(in efficiency) scoring numbers.

People might look at that Miami team falling apart(getting swept by the Bulls in 2007) and put it on him, but that whole team was old - Mourning and Payton were about done, Antoine Walker and Jason Williams had fast declines, etc. There's a reason they pretty much got rid of everyone other than Wade(and Haslem).

I think it was only those last two years with Cleveland and Boston where he really looked done.

Offensive Gap vs Defensive Gap

As mentioned in the previous section, Duncan's career average TS Add over nineteen seasons is 53.1. Shaq's career average TS Add over nineteen seasons is 145.2.

Duncan recorded a 4+ RS rTS three times in his career, all of them in those first six seasons. Shaq recorded a 4+ RS rTS fifteen out of his nineteen seasons.

Shaq never recorded a negative RS rTS, Duncan did it four times.

Shaq's advantage over Duncan as a scorer is not a small one. It is a large one.

On the other side of the court, Duncan is clearly the superior defender; he was elite. But that doesn't mean Shaq wasn't solid. I feel like the numbers would characterize him as good-but-not-great. Just by his sheer size(not just height, but his girth too) and shot-blocking ability he was going to have an effect. It was like a wall in front of the basket.

Shaq had a 2+ D-RAPM eight times in his career(and that is without considering his Orlando tenure, which came before on/off was tracked). This obviously falls well short of Duncan's 15 3+ and 9 4+ D-RAPMs, but it's also far from nothing.

Take the Lakers' DRtgs for the years Shaq, Kobe, and Phil Jackson were together:

Code: Select all

1999-00: 98.2(1st)
2000-01: 104.8(21st)
2001-02: 101.7(7th)
2002-03: 104.7(19th)
2003-04: 101.3(8th)


That's three top ten finishes and two around the border between the middle and bottom third. Now look at Shaq's D-RAPM in those same five seasons:

Code: Select all

1999-00: 2.31
2000-01: 0.7
2001-02: 2.49
2002-03: 1.14
2003-04: 2.74


Shaq's higher D-RAPMs seem to correlate with the team's higher DRtgs. This would seem to suggest he did have some level of actual defensive impact(as well as his inconsistency on that end, to be fair), at least during his peak years. And FWIW, those 2+ D-RAPMs didn't end when he left LA; he posted 2.07 in 04-05, 2.27 in 05-06, and 2.44 in 07-08. After 2008 is where his defense fell off a cliff.

My point with all of this is this: Shaq, while clearly not the defender Duncan was, was still a solid defender, and because of this I believe Duncan's defensive edge over Shaq can be matched by Shaq's sizable scoring edge over Duncan.

There is also the issue of playmaking, where Duncan also has an edge. Looking at bbref's "points generated by assists" stat(by no means the definitive metric for this, just one example) and taking their career averages in the regular season and the playoffs:

Duncan RS: 496.15
Duncan PO: 97.8
Shaq RS: 352.86
Shaq PO: 75.78

This is just one illustration of Duncan's edge in this department. What I will say for Shaq though, is that it seems logical to me that if Shaq is scoring more points than Duncan on average(Shaq averaged 35.2 points per 100 possessions for his career in the regular season and 34.7 points per 100 possessions for his career in the playoffs; Duncan averaged 29.7 in the regular season, and coincidentally, 29.7 in the playoffs), then that means he'd be creating less points for his teammates than Duncan because you only touch the ball so many times. So you can view this playmaking gap as a detriment to Shaq's game, or as the logical result of his dominance as a scorer.

Multiple Contexts

I bring this up because the "multiple contexts" argument was used a whole lot by many of you when arguing for LeBron, so I feel like it ought to apply here too. Shaq and Duncan have similarly glowing resumes. Duncan never had a <50 game winning season, he went to the finals six times, and won five titles. But he was drafted into an ideal situation, a team that had been in the WCF just two years earlier, and played his entire career for one coach.

Shaq won 50+ 16 out of 19 years.

Shaq came into the league with Orlando, who were an expansion team going into its fourth year of existence having never won more than 31 games, and spearheaded a 20-game(21 to 41), nearly 8-point SRS swing(-6.52 to 1.35) in his rookie year. By his third year(yes, with some crazy 1993 lottery luck), they were in the Finals.

He went to LA at a time when they were just starting to get good again after Showtime and anchored an era that saw four finals appearances, a threepeat, and eight straight 50+ win seasons.

He went to Miami, where the Heat had been a 42-win, -0.13 SRS team that sort of surprised in how hard they pushed the Pacers in their six-game second round series, and, along with Wade, was at the center of a 17-win, nearly 6 SRS point swing(and that's with them losing Lamar Odom and Caron Butler in the trade) that saw them get one game away from the Finals(and a lot of people still think they would've gotten there and possibly won it all if Wade hadn't gotten hurt in that series).

Shaq had a stretch of his career where he won at least 50 games twelve times(they were at a 50-win pace in the 99 lockout season) and went at least to the Conference Finals nine times(including six trips to the Finals and four championships) in twelve seasons across three different organizations with three different second options, at least three distinct supporting casts(maybe more than that considering how the lineup around him and Kobe kept changing from Eddie Jones/Nick Van Exel/Cedric Ceballos to Glen Rice/AC Green/Ron Harper to Robert Horry/Rick Fox/Derek Fisher/Horace Grant to Karl Malone/Gary Payton/Devean George), and five different head coaches(Brian Hill, Del Harris, Phil Jackson, Stan Van Gundy, Pat Riley).

Shaquille O'Neal strikes me as one of the safest bets to build a contender around in the history of the NBA.

As the final thing I'll say about Shaq/Duncan, going purely by on/off(note again that Shaq's numbers don't include his Orlando years because the data wasn't tracked):

Shaq's career average RS on/off: +8.5
Duncan's career average RS on/off: +8.0
Shaq's career average PO on/off: +11.7
Duncan's career average PO on/off: +8.5

A 3+ point on/off advantage in the playoffs is something worth considering.

I also lean towards Wilt over Duncan for a similar set of reasons. Wilt and Shaq were very similar, except that Wilt was a better defender and had less team success(at least in part due to some bad luck - poor roster with the Warriors, no Billy Cunningham in 1968, Breda Kolff wouldn't put him back in + improbable Don Nelson shot in 1969, no Jerry West in 1971, no Happy Hairston in 1973). I may have more to say about Wilt later, but I probably can't make a better case for him than ZeppelinPage just did.

Right now, I think I've got Wilt/Shaq, and then Duncan/Hakeem(exact order not determined yet). And I may put Magic before Duncan/Hakeem too.

I mean, you are arguing for Duncan's longevity being overstated by looking at Shaq's (not Duncan's) biggest strength that is scoring, but even though you noted that Duncan's defensive longevity absolutely crushes Shaq's, you say that is caused by Duncan being a better defender... Sorry but I find it inconsistent.

I don't think it is inconsistent. I am simply arguing that

A)Duncan and Shaq each have a part of their game that fell off - Duncan's volume scoring and Shaq's defense - and other parts that remained effective until the end - Duncan's defense and playmaking and Shaq's efficient volume scoring - and

B)That those arguing for Duncan's longevity over Shaq's have a way of treating Duncan's decline as a scorer with kid gloves while going hard at Shaq's late-career defensive deficiencies.

Duncan has much worse scoring longevity, because he's much worse scorer. It's that simple. Even at his peak, Duncan never touched Shaq's scoring production. It's easy to understand why Duncan's scoring aged worse than Shaq when you take that into account.

In comparison, Shaq's defensive "longevity" (above average defensive seasons) are probably less than q quarter of Duncan's (which is his whole career basically).

See, I don't agree with either characterization here. Like I said in my post, Shaq's career average TS Add is 145.2, and Duncan's average TS Add for his first six seasons is 122.5. In that early part of Duncan's career, he wasn't nearly as removed from Shaq as a scorer as he'd become.

And I suppose it would depend on how you're defining "above average defensive seasons", but out of the fifteen Shaq seasons we have tracking for, he had a 2+ D-RAPM in eight of them, which is certainly more than the "quarter" you suggest, and if we had data for the Orlando years, it could possibly bring it up to "half". No one is arguing that Shaq was ever the defender Duncan was, but let's not give him short shrift either.

By this logic, you could argue that Bill Russell was done in 1968 because he became a below average scorer at this point, when in reality he was still in conversation for the best player in the league.

At no point in my post did I ever say Duncan was "done", only that one aspect of his game fell off.

I notice you didn't address the other arguments in my post - different contexts, playoff +/- gap, etc.

At any rate, Duncan's clearly got the #5 spot locked up, the vote doesn't seem particularly close. I just don't agree with it and I frankly amazed at how strongly this board views him as top 5 guy.

[ZeppelinPage]

Vote: Wilt Chamberlain
Alternate: Shaquille O'Neal
Nomination: Jerry West

Shaq gets the nod as the alternate vote for me. Dominance in the playoffs made him seemingly unstoppable at times.

With Curry already getting some nomination votes I'd like to get Jerry West in the discussion here soon. Incredible two-way player and ranks 10th in Career TS Added on high volume.

Here are some posts I've made during this project to explain my Wilt vote:

1. Wilt Chamberlain's Personality: Being Coachable & His Desire to Win

2. Wilt's Playoff Luck Regarding Injuries

Spoiler:

Wilt wasn't as fortunate to have a consistently healthy and deep team, which is why I don't think it's as simple as assessing his championships and blaming him. If a player performs in a playoff setting but his teammates falter, I don't hold him accountable unless there's reason to believe otherwise. After all, basketball is fundamentally a team game played by five players. Wilt faced the most formidable competition of any superstar in NBA history throughout his career, always performing at or above his regular season level that his teammates, either injured or floundering, couldn't match. Much of the time, Wilt was losing to teams with a better SRS, the teams usually being apart of the greatest dynasty in NBA history. Therefore, he was regularly confronted with a steep mountain to climb.

Not only were his teams often the underdogs, but his teammates frequently grappled with injuries. Here are some examples:

1962: The Warriors lose on a last-second Sam Jones buzzer beater in Game 7. Tom Gola, their second-best player and exceptional defender, was essentially out for most of the series, playing only 107 minutes in 4 games. Wilt came close to defeating the Celtics here.

1965: Havlicek stole the ball. Larry Costello played through injury the entire series and averaged only 5 points per game. Another extremely close game was played without a key player.

So, his team's lack of talent was further exacerbated by injuries.

Looking at '68-'73, almost every post-season besides 1972 involved some kind of injury:

1968: Almost the entire starting lineup was injured and missing Billy Cunningham.
1969: Jerry West had a torn hamstring, and van Breda Kolff didn't put Wilt back in the game.
1970: Wilt returned early from a knee injury, and Jerry West played in the Finals with injuries to both his hands.
1971: Jerry West was out for the playoffs.
1973: Both Wilt and West were injured.

I'm not sure how much blame I can place on Wilt when, in many instances, a key player was either out or playing injured. These injuries are beyond Wilt's control. In the playoffs, he gave his teams a substantial boost, trying to overcome factors beyond his control like injuries and roster construction.

This is a rather drawn-out way of explaining why I don't center my arguments around championships, or the lack thereof, when determining a player's impact. A player can only play with the hand he's dealt and do his best to overcome adversity. Despite Wilt being plagued with untimely injuries and lackluster teammate performances, I believe he offered his team an unparalleled overall boost in every area of the game.

3. The Complexity of Comparing Wilt's Game to Modern Eras

Spoiler:

True, 70sFan. Not only that but I also think comparing Wilt's game with modern eras is a complex matter.

When analyzing these players it's important to think of the game without our benefit of hindsight. We can look back with a modern lens but they could not. A lot of rules that aided future players were not available to earlier eras.

For instance, dribble rules and overall spacing made finding cutters and anything other than jumpers significantly more difficult in the 1950s and 1960s. Refs were also harsher on travel calls. This is why the fast break was so popular as it was the easiest way to score with the spacing it provided before the defense was set. It was also more difficult because, as you know, the offensive foul rules and quick three in the key calls that refs would make. Creating your own offense in the half-court was not an easy thing to do back then and teams relied more on moving the ball around.

This leads into why Hannum implemented his "wheel offense" with Wilt, where players would constantly move around him in a wheel to keep the defense guessing and give Wilt options as a way to counteract how difficult it was to score in the half-court.

There is a substantial difference in the rules between what Wilt and guys like Jokic and Jordan played under. The rules in Wilt's time made it harder for a player to back a defender down and apply pressure and "generate offense" consistently.

I don't really care how Shaq played in 2000s, I don't care how Jokic is playing, I don't care how Hakeem or any other center outside of his era played because, to me, it doesn't matter. Wilt could only play within the confines of his rules and era. And within that era, with him playing that way, the team was the greatest offense ever. His passing and scoring must have been really valuable.

Although there are certainly instances where Wilt draws doubles and passes or hits cutters on film. I just don't find it fair to watch Jokic or someone else and then criticize Wilt for not playing similarly, because they played under separate rules where the game was thought of in different ways.

[70sFan]

Well it's still 14th out of 18. That's still much nearer the bottom than the top. Curry to Kareem is also a sizable jump in win percentage.

Yeah, he's also 2nd in the same list in terms of volume of such series played. Curry is "better" by winning 12 series less and losing 5 series less.

To be honest, I don't find this argument persuasive at all, but it's possible I miss something.

And when I go back and watch games (do you have a source better than youtube because the selection is limited?),

Yes, I do, but I don't have the access for them right now (I am out of my country). I will share them for you next week.

the pattern always seems to be what I remember. Duncan eats early, then Shaq gets put on him in the 4th, then we don't hear from Duncan (and I wanted the Lakers to lose).

Again, I already posted the video but I will post it again:

Watch on YouTube

ISO scoring:

G1: 3/6 FG, 1 ast, 1 tov, 0 fouls drawn
G2: 0/0 FG, 0 ast, 0 tov, 0 fouls drawn
G3: 2/4 FG, 0 ast, 0 tov, 1 fouls drawn
G4: 0/2 FG, 1 ast, 0 tov, 2 fouls drawn
G5: 1/5 FG, 2 ast, 2 tov, 2 fouls drawn

Game 5 I had Duncan at something like 2/9 in the 2nd half when Shaq was his primary defender (you seemingly have different numbers).

I have him 3/11 in the 2nd half when Shaq was his primary defender, not a big difference.

The thing is that game 5 performance isn't representative at all and I think you got his conclusion only from this game.

I just watched Game 4 for Duncan/Robinson guarding Shaq so I didn't track Shaq guarding Duncan, but Shaq blocked him I believe 3 times straight up because Duncan kept trying to turn right and shoot with his right and was practically just throwing it into Shaq's arm.

Could you show me these plays?

You say you don't have Shaq guarding him much but do you have a breakdown by quarter? Because I certainly don't think Shaq did much in the 1st half or early 3rd.

Just from my video I made, Duncan's FG/FGA when he was guarded by Shaq (so excluding help defense):

G1:

0/0 in 1st
4/4 in 2nd
1/2 in 3rd
1/3 in 4th

G2:

0/0 in 1st
0/0 in 2nd
1/1 in 3rd
0/0 in 4th

G3:

0/0 in 1st
0/1 in 2nd
1/3 in 3rd
1/2 in 4th

G4:

0/0 in 1st
0/1 in 2nd
0/0 in 3rd
0/1 in 4th

G5:

1/1 in 1st
0/1 in 2nd
1/5 in 3rd
2/6 in 4th

Again, this is a clear picture of what I've been saying in a long time - game 5 isn't a good representation of what happened throughout the series. I understand that Shaq finishing strong in game 5 may create a narrative, but we should focus on more than just narratives in this project.

And I think it's why the narrative from those years was that Shaq was better than Duncan. Because Duncan didn't seem to take it to Shaq directly. And while Duncan might have played very well in 2002, getting just straight up annihilated in 2001 stuck in people's minds (and should have) and 2004 was a straight up poor series from Duncan.

Well, Shaq didn't do any better than Duncan in 2002 series and he got completely outplayed in the next year. Not to mention 1999...

They have very similar stats (at least by box score) over their playoff careers. I'm voting Duncan 6th. He obviously has performed well. But my central thesis is that Hakeem didn't just perform well, but seemingly always performed well in his biggest moments, the ones where a championship might at least be reasonable (even if still quite unlikely), and these were moments where success was not a foregone conclusion, either because his team was an underdog or his counterpart on the other team was often viewed as an equal or even a superior until Hakeem ended that conversation decisively.

While the consistency of Hakeem's performance under those circumstances was indeed fantastic and worth mentioning, we're still comparing someone who had a brief window of championship possiblity at his best vs someone who competed for titles for the majority of his career. I think comparing the quantity of "good and bad" series between them is pointless with such a different environment they played in.

My list wasn't just 4 series where Hakeem played well, but all came from the conference finals or later, and all with all-time players on the other side (i.e. not his matchup with Joe Kleine). Hakeem did not get 9 or 10 conference finals or 5 or 6 finals to have great moments, with a few chances to slip up here or there and still win some championships. He got 4 conference finals and 3 finals. From my list, in 4 of those he massively outplayed a great, 3 of whom of were centers. And it's not like the other 3 were bad. He was easily the best player in the 1994 WCF and outplayed Shaq at least a little in the 1995 Finals. Only in the 1986 Finals, where he got bested by peak Larry Bird while still putting up 25/12/3 could you say he wasn't the best player. And the Rockets still took 2 games off a Celtics team that had just monkey-hammered a +8.7 Milwaukee team by sweeping them with an MOV of 15 ppg.

That's 7 Series featuring:
4 Top 10 Players (Magic, Kareem, Shaq, Bird)
3 Top 20 Players (Malone, Robinson, Malone)
3 Top 30 Players (Stockton, Ewing, Stockton)

Each series had at least 1 of those players and 3 had 2. 9 of those guys were in prime seasons, probably half at peak (or very near) seasons, and Kareem was good enough that he was 1st team All-NBA.

And he outplayed 9 of 10 players and won 5 of the 7 series while usually being a significant SRS underdog (except 1994 Utah).

What 2nd Malone are you talking about?

Well, you can create a similar list for Duncan:

1998: Malone
1999: Garnett, Shaq, Kobe
2001: Garnett, Dirk, Shaq, Kobe
2002: Shaq, Kobe
2003: Shaq, Kobe, Dirk
2004: Shaq, Kobe, Malone
2005: Nash
2006: Dirk
2007: Nash, James

Out of these 10 seasons, I counted 14 such series. We can count out Garnett series because Minny were overwhelmed by Spurs talent, so 12 such series. In these 12 series, Duncan outplayed Shaq (top 10 all-time) 3 times (1999, 2002, 2003), LeBron (top 10 all-time) once (2007), Kobe at least two times (1999, 2003, likely 2003 as well), Malone 2 times (though we shouldn't really count 2004), Dirk all 3 times (with 2006 being arguable) and Nash 2 times. Again, weaker series happened more often but he has way more such series.

I think the narrative from the 1994 Finals was that Hakeem and Ewing were both solo stars and Hakeem outplayed Ewing so he won. But I would say the Knicks had the better supporting cast. They were actually a 60 win team by SRS compared to Houston at 53 wins. Only luck/randomness (clutchness?) got the Rockets homecourt in the first place at 58 wins to 57 wins. This is a series where the Knicks actually outscored the Rockest by 0.7 points per game. Where every game was decided by single digits. Hakeem can't have a 2005 Finals where he just can't score against a great frontline and all-time defense. He can't even just outplay Ewing. He has to destroy him to just eke out a game 6 and game 7 victory. And he does.

I won't try to take away anything from Hakeem's performance, but I don't think it's that clear the Knicks were more talented. I think Riley did a tremendous job at slowing down Rockets offense and despite his horrible shooting, Ewing did a tremendous job defensively as well.

I think the narrative from the 1995 WCF was Hakeem destroyed Robinson and the Rockets cruised to the Finals and beat the Magic. But the Rockets didn't cruise. They outscored the Spurs by less than 2 ppg. They didn't win while Hakeem destroyed Robinson. They only won because he did. Game 1 is a 1 point game where Robinson shoots 5-17. Game 6 is a 5 point game where Hakeem puts up 39/17 while Robinson shoots 6 for 17. Hakeem can't just outplay Robinson if he wants to advance, he needs one of the greatest playoff series of all time. And he delivers it on cue.

Well, I think you fixiate too much on the averages in this case. Spurs were that close in average score only because they won game 4 by 22 points. Without this blowout, the Rockets won comfortably all but first game.

There's no 2004 series loss vs the Lakers where victory just required a typical Tim Duncan series, not some all-time great series.

I'd say the Rockets could do a lot of damage in the playoffs in 1996 had Hakeem played up to his standards against the Sonics. That's a comparable situation and no, Duncan didn't have a better team around him than the Lakers.

I bring up expected titles, because Duncan's 5 titles against 4 expected basically just make him look like a guy who got 19 chances to play with good teammates and great coaching and converted them about like you would expect. Overperform some times, underperform some times. It all averages out so you'll still get your 5 titles. Lose as a #1 SRS? Well, just come back next year and be #1 SRS again. Have a bad 2004 series, it's not the end of the world. Your team will still be great in 2005 and Ginobili will go crazy in the playoffs. Lose a heartbreaker in 2006, there's always 2007. Have a down 4 years, don't worry the front office will reload.

Ok, but these numbers don't give you any idea about how each player played or what's the circumstances of these losses. That's why I don't see this as a very valuable approach.

Hakeem's numbers won't change because that's when all those series are from.

Duncan didn't play a ton in that time frame and goes to 3-3 (including 2006 Dallas).

So 6-6 vs 3-3.

Thank you, so Duncan's numbers got noticeably worse when he went past his prime. That could suggest that the Spurs needed him more than some may want to admit.

Maybe that's not what you were doing, but I always get a little squeamish when longevity guys start not getting later career things counted like Kareem getting outplayed by Moses in 1981 and 1983 or Hakeem in 1986.

I count everything, but I also try to contextualize it.

Kareem wasn't past his prime in 1981 (he was later) and he wasn't outplayed by Moses in that year, but that's another discussion.

No. Is there a way to get that? Robinson had a good split between off and on minutes (35 to 13 per game) and also was +20 on and -15 off, so this wasn't just a +3 on, -32 off situation due to an ill-timed 12-0 run by the other team.

I don't know where to get such numbers from 1999. Consider this though (numbers from basketball-reference):

Duncan and Robinson: 531 minutes, +19.3 ON
Robinson: 598 minutes, +20.3 ON, -14.8 OFF
Duncan: 728 minutes, +10.4 ON, +14.0 OFF

Robinson played total of 67 minutes without Duncan in the playoffs, which gives us less than 4 minutes per game. In comparison, Duncan played almost 200 minutes without Robinson, which gives us 11.6 mpg to be precise. Duncan basically played full quarter without Robinson, while Robinson played almost all minutes next to Duncan.

Sure, they were obviously good. But also good without duncan. I'm pretty sure the adjusted stuff still shows Duncan as good, but it's still quite an odd thing for someone to be a straight up on/off negative guy as the leader of a title team. Only 7 of the 49 guys I looked at were negative, and other than 2001 Shaq, you have Duncan a whopping 3 times and the other 3 were all secondary players (1998 Pippen, 2006 Shaq, 2013 Wade) and almost all of those guys are viewed as struggling in those playoffs. when your case is built on being invaluable, winning 3 championships where your team was doing even better without you on the floor should at least be something to discuss.

The case of 2001 Shaq should already tell you something about such a criteria.
As for 2014 Duncan - yeah, he wasn't "invaluable" in a way that you couldn't replace him with anyone else in the league. This title was definitely the case of Duncan being a part of a very strong team around him - which is why nobody mentions it as a serious accomplishment that should put him ahead of someone like Hakeem.

Mind you that we don't have the on/off numbers for Hakeem, so we don't know how he'd look by them.

Thanks, look forward to more back and forth.

You are welcome!

[70sFan]

I don't think it is inconsistent. I am simply arguing that

A)Duncan and Shaq each have a part of their game that fell off - Duncan's volume scoring and Shaq's defense - and other parts that remained effective until the end - Duncan's defense and playmaking and Shaq's efficient volume scoring - and

B)That those arguing for Duncan's longevity over Shaq's have a way of treating Duncan's decline as a scorer with kid gloves while going hard at Shaq's late-career defensive deficiencies.

OK, but the problem is that:

A) Duncan never fell off as a scorer to the point when he became a liability (well, at least until his last season), while Shaq's defense was very concerning way earlier than in his 19th season,

B) I don't think you can argue that Shaq's scoring ability made him comparably effective player as Duncan's defense late in their careers. Shaq was basically done as an impactful player after 2006 (his 14th season), while Duncan kept coming after his 14th season (which was 2011).

See, I don't agree with either characterization here. Like I said in my post, Shaq's career average TS Add is 145.2, and Duncan's average TS Add for his first six seasons is 122.5. In that early part of Duncan's career, he wasn't nearly as removed from Shaq as a scorer as he'd become.

Yeah, but that's only because Shaq missed a lot of games in his career and Duncan didn't miss much games in his first 6 seasons. If you look at TS+, Duncan's best season would be at the end of Shaq's best 10 seasons.

And I suppose it would depend on how you're defining "above average defensive seasons", but out of the fifteen Shaq seasons we have tracking for, he had a 2+ D-RAPM in eight of them, which is certainly more than the "quarter" you suggest, and if we had data for the Orlando years, it could possibly bring it up to "half". No one is arguing that Shaq was ever the defender Duncan was, but let's not give him short shrift either.

What database are you using? Englemann's database doesn't show 2+ DRAPM in 8 seasons, not even close to that.

At no point in my post did I ever say Duncan was "done", only that one aspect of his game fell off.

OK, but my analogy with Russell still works here. Yeah, Duncan's scoring fell off, but his impact didn't. Shaq's scoring didn't fell off, but his overall impact did.

I notice you didn't address the other arguments in my post - different contexts, playoff +/- gap, etc.

Yes and I apologize, I don't have enough time to adress everything - I am at the scientific conference outside my country this week.

[DraymondGold]

Voting Post
Vote: Tim Duncan
Nomination: Steph Curry

It looks like Tim Duncan's likely to win this (which I agree with -- more on that in a sec), so I'll cast my vote for him too. Looking ahead, I think my next group will be: Wilt, Magic, Shaq, Garnett, Curry. I'm willing to be convinced on the order of these players still, pending more statistical analysis from me and hopefully continued great comments from everybody else. Looking forward to it!

Brief reasoning for Duncan: I see him as the best combination of having a good statistical profile, with a reasonably high peak and arguably the best longevity of these next players (Garnett?). His offensive scalability and verstatility is good *enough* to get him over the other guys (though I might take some of the other guys), and the team-level results are obviously great.

Statistically: As have been discussed over the past few days (and some previous threads), Duncan is one of the top players in the basic plus minus data.
-He has the highest career raw on/off of the available players (Duncan +8910, Curry +5689, Garnett +4913, Shaq +4887).
-He looks great in Goldstein's (full season?) RAPM: 4 years of 7+ RAPM (Garnett has 6 but that includes 2009 which has lower minutes/games played; Shaq has 4, Curry has 2 without 2016). He also seems to have aged better in RAPM than Shaq, though again Garnett looks better in Goldstein's sample. Not all RAPM sources agree perfectly though, and I believe Duncan closes the gap with KG in the playoffs (at least somewhat).
-He has the best career PIPM "Wins Added" (PIPM per 100 possessions x total possessions played, divided by some threshold set as the number of PIPM to get a win on average): (Duncan 284 wins added, Garnett 261.4, Shaq + 232 (with box estimates for early years), Magic + 188 (box estimate), Curry +142 (pre 2021, ~181 if we add 3 average career years, ~202 if we add 3 average prime years)
-WOWY-wise, he has less of a clear advantage, but I've made the case that raw WOWY has large uncertainties & underrates players with better coaches. In total career WOWYR (career average WOWYR per game * total games), he looks better (Magic +9060, Duncan +6542, Garnett +6433, Shaq +6276, Curry unknown but looks GOAT level in raw WOWY).

Peak wise, I have him 7th (below Wilt, Curry, Shaq). So it's definitely the longevity that ends up helping him out.

Team-wise, he's pretty clearly up there with these guys, although some of this credit goes to Popovitch/Manu/Parker/Kawhi/etc. I don't have nearly the same team building concerns with Duncan as I do with Hakeem. Ideally his teams would show a bit more offensive oopmh during his peak/prime -- a bump in Duncan's passing, interior gravity, or off-ball distance shooting would likely help here -- and these sorts of offensively impactful / scalable skills are areas where the other players in this tier catch up. However I wouldn't exactly call Duncan a negative per-se: he clearly scales well on defense, he shows some versatility in his offensive style in his shift from post/iso-centric offense to be a bit more of a passer and screener. Not as much as Wilt, but not nothing either, and this is also helped by his positive contributions to team chemistry/leadership/locker-room.

...

As for my nomination for Curry, I spoke about the reasoning with jake over the past two threads. Curry's statistical profile is crazy good, and even though he doesn't have the longevity of some of these guys, he's ahead by enough of a margin in some of these stats (on/off, RAPM, WOWY, etc.) that one can still argue for Curry from a career perspective.

Next time: Wilt vs Magic vs Shaq vs Garnett vs Curry!

Discussion questions for certain players:
-Wilt: Wilt's a guy that has a large uncertainty range for career ranking. I have his peak as top 5 (over Shaq and Garnett and certainly Hakeem), but I also see him as far more inconsistent year to year in terms of how much he helped him win. Some of this is due to Wilt's GOAT-level versatility (Pro!) but sometimes e.g. in 63 or 65 I think Wilt's chosen style / effort didn't maximize his team's chances at winning the playoffs (Con!).
Question: When Wilt decided to pursue a certain style, how much of these decisions were coaching vs self-compelled? I've seen quotes around (specifically the 70s "play like Russell" one), but I'd love to see a clear connection drawn between Wilt's *style* and either his or his coach's goal for the season.
Question: If we allow that Wilt's style wasn't always perfect for winning (I'd argue his balance of scoring / rebounding / playmaking / defense in 64 and 67 for example were better than in 63 or 65 for example), how much can we attribute this to the knowledge of the era (or of his coach). If Wilt were have better coaching (say Alex Hannum for longer, akin to Duncan having Pop, Shaq having Phil, Curry having Kerr for longer), what can we reasonably expect from Wilt's style, stats, and team performance?

[eminence]

I don't have Wilt as valuing good coaching, he voluntarily left Hannum to go make movies.

[ceoofkobefans]

Vote: Hakeem Olajuwon

This is a big shift for me and easily the most I've changed my mind in one discussion period. I always want people to be open to my arguments though and so I feel it's only fair to be open to changing my mind when presented with compelling evidence. After being initially impressed with f4p's post, there were some things that I investigated myself:

-First, Hakeem did have significantly better stats than Duncan in the playoffs throughout his career. His edge in the postseason was similar to Duncan's edge in the regular season. Hakeem had a career BPM of 6.9 in the playoffs compared to 5.9 for Duncan and this advantage holds even if you use an age cutoff to keep Duncan from having so many playoff games past his prime, late in his career.

-Both WOWY and WOWYR data give Hakeem the edge and there's a significant sample for both. For instance, when Hakeem missed the playoffs in 1992, the Rockets went 40-30 with him and 2-10 without him. This is an even wider split than Kareem had in 74/75 when the Bucks missed the playoffs and is very understandable. Hakeem didn't miss enough games for it to be a problem, but he missed enough to see that yes, most of the time, he did have a very poor supporting cast in Houston and I can't really blame him for failing to elevate his team consistently to the levels that Duncan did.

-I also feel relatively confident that Hakeem had a significant defensive edge. He didn't play on as good of defensive teams as Duncan and didn't always put up eye-popping rDRtg numbers on a team-wide basis, but I do feel that I would have to grade him as a meaningfully better defender overall based on multiple educated opinions as well as the few data points that we do have.

-I don't feel that Duncan had a meaningful edge in longevity. Duncan played 19 seasons to Hakeem's 18 and who had more meaningfully impactful seasons kinda depends on where the cutoff is for meaningful. If you just take their top 10 seasons away, Hakeem might gain more in seasons 11-13 than he loses in seasons 16-19 is 14 and 15 are roughly equal.

So when it comes down to it, I'd say:
Scoring: Hakeem > Duncan
Passing: Duncan > Hakeem
Defense: Hakeem > Duncan

And there's no way the difference in passing is enough to overcome the other two. Both did an absolutely A+ job of being clutch winners within the constraints they played in and both were a near push on longevity. I think I have to give Hakeem the edge as the superiorly skilled player by the slightest of margins.

Alternate: Tim Duncan


Nominate: Kevin Garnett: Tremendous player, analytics darling, and honestly, the one other player I could possibly be convinced to put ahead of Tim Duncan in this project. Somehow manages to show out as somewhere between the 2nd and 4th most valuable player of the past 25 years in pretty much every dataset despite being used completely wrong as a defender for much of his Minnesota years, taking away from an all-time strength as a help defender to position him on the perimeter.

Voting for Hakeem and having Kobe not top 20 all time makes 0 sense lol

[lessthanjake]

I don't have Wilt as valuing good coaching, he voluntarily left Hannum to go make movies.

I think that’s right to some degree. Wilt also did get almost twice as much money from the Lakers, though, and the opportunity to do things in LA that would make him more money. And that surely mattered quite a lot in a context where the money Wilt was making, even after almost doubling his salary at LA, was less than the veteran’s minimum now (in inflation-adjusted terms). Playing basketball was not the kind of money-making venture it is now, such that someone couldn’t really afford to leave other financial opportunities on the table. Not that this really matters for purposes of how good a basketball player he was, except just that I think his willingness to leave a good coach was reasonable given the financial context of the NBA back then.

[Doctor MJ]

So, in the name of longevity evaluation, I'm going to post the guys who made my POY, OPOY & DPOY ballots the most going back into history. I don't want to be coy here: Magic & Curry are on my mind, and these numbers will show you why.

Before I list them, I feel I should link to my spreadsheet that gives you year-by-year details, but I feel a bit self-conscious about it. I really should have taken notes as I went through and did this so I could at least answer the question of "Why that there?" and give people a productive place to try to augment my vision, but I didn't. That means that while people can ask me how I came to certain conclusions, the reality the best I can do is try to evaluate the thing again in a lot of cases, and I may not have time to.

Really I'm posting this stuff not to convince you that my perspective is right, but just to try to give simple data to understand roughly where my perspective is. I don't think there's any doubt that any of you who do the same thing I did would get different results, and sometimes vastly so.

My thought is that where you know you differ, that can easily explain the differing conclusion in the context of the 100.
And of course, if you don't differ all that much, but do have very different conclusions, that may provide the opportunity for minds to be changed.

Here we go. Most years getting POY shares (Top 5 player) from my far-from-perfect perspective:

1. Kareem Abdul-Jabbar 13
(tie) LeBron James 13
(tie) Bill Russell 13
4. Wilt Chamberlain 11
(tie) Magic Johnson 11
(tie) Michael Jordan 11
7. Oscar Robertson 10
8. Steph Curry 9
(tie) Jerry West 9
10. Larry Bird 8
(tie) Tim Duncan 8
(tie) Kevin Durant 8
(tie) Karl Malone 8
(tie) George Mikan 8
(tie) Dirk Nowitzki 8
(tie) Bob Pettit 8

OPOY shares (Top 3)

1. Kareem Abdul-Jabbar 12
2. Steph Curry 10
(tie) Magic Johnson 10
(tie) Oscar Robertson 10
5. Michael Jordan 9
6. LeBron James 8
(tie) Steve Nash 8
8. Shaquille O'Neal 7
(tie) Jerry West 7
10. Wilt Chamberlain 6
(tie) Kevin Durant 6
(tie) Karl Malone 6
(tie) Dirk Nowitzki 6

DPOY shares (Top 3)

1. Bill Russell 13
2. Wilt Chamberlain 9
(tie) Hakeem Olajuwon 9
4. George Mikan 7
(tie) David Robinson 7
(tie) Nate Thurmond 7
7. LeBron James 6
(tie) Michael Jordan 6
(tie) Slater Martin 6
10. Tim Duncan 5
(tie) Patrick Ewing 5

Before I get into specific takeaways, I'll note that doing POY-style analysis has eventually re-shaped the criteria I'm using in this project. I like going year-by-year because I like trying to immerse myself in a given season. I feel like it drives me to keep learning more about basketball history.

It has the drawback that I don't then that I really didn't feel like potentially nixing out entire swaths of historically successful players based on 2020-vision, which means that my side-by-side comparisons become less basketball-real, and so it's not the only lens I want to use...but I thought I could apply it most consistently in this project which would make communication more clear and experience more...say, additive rather than subtractive. Time will tell if this is right.

But one of the things I also really appreciate it for is the longevity-view of the Top 5 (or 3) threshold. I track other stats that do this in different ways that I like to see how they count up, and when you're looking at different levels of players, different things work as longevity factors.

When viewing it from a GOAT perspective among guys who truly dominated the league, I think the POY lens is pretty meaningful. How often do I really think a guy who has a lower number here has better-enough longevity that it would make up for what seems to be a lesser prime?

So, the starting point here is Russell. If no one has more Top 5 seasons than him, because he made it every year of his career and nobody else who has played more than that has qualified more, how comfortable am I using longevity as the deciding factor in my list? Not very. I'll agree that LeBron & Kareem have superior longevity, but Russell was great enough for long enough that it's just really hard for me to factor it in that strongly.

Alright, more thoughts:

* I think Kareem, and Kareem's offense specifically really stood out. Did arguments made on Kareem's behalf in recent years influence me in this pass? Oh absolutely. And yes, I considered voting for Kareem ahead of Jordan. Didn't happen this time. Could happen in the future though.

* Magic stood out in a closest career-to-Russell-since-Russell way. By the NBA's recent awards rule change of game playing thresholds, Magic would qualify in 11 seasons. He made my POY ballot all 11 of those season...and as you can see, not a lot of guys did more than that that.

* Wilt did better than I expected. This doesn't wash away all the concerns I have about his career, but it does make them loom a bit less large.

* Curry's presence here might be the thing that goes furthest again what many knowledgeable people believe. Many see Curry as a guy who hasn't been great for that long, and as you can see, I think he's been great a pretty long time. And yes, I too was surprised when I ended up with Curry making my OPOY more times than LeBron. Obviously this is a situation where if I were tallying Top 5 or Top 7 instead of Top 3 it would change things - which also pertains to why another man's list would look very different.

To be clear: I'm not looking to say that Curry should rank higher than LeBron on an Offensive GOAT list, and this is not any direct statement about Peak it's just that when I went through year-by-year, this is how the count turned out.

* Over on D: I think the thing that might aggravate people the most is having Duncan only there with 5 DPOY ballot instance...below both Jordan & LeBron with 6. Analogous to the O situation from before, I have Duncan ahead of Jordan & LeBron by defensive career, but this is how it came out.

* Before any one responds to any of these points, I want to end this directing your attention to who is missing from this list entirely y'all know me as a champion of:

Kevin Garnett

For the record he makes my POY ballot 6 times, my DPOY 4, and my OPOY once. The most telling thing I can say is that Garnett's 4 DPOY ballot appearances come once he gets to Boston in his 13th year in the league. Do I think he was incapable of doing this until his 13th year? No, I honestly think he could have been doing his Boston thing by his 2nd or 3rd year if the rules then had allowed Thibs scheme. In a metric where only Bill Russell reached double digits, I think in another universe Garnett gets there too.

And yet, Garnett is falling in my 100 vote this time because of the criteria I'm using. Frankly makes me sad, and more than anything else I want people to know:

Kevin Garnett was a freakishly insane basketball talent who was recognized early on as an insane talent but whose freakish nature - relative to norms at the time - held back his cumulative achievement. That's why people underrate him. That's why my own present ranking will underrate him relative to good he actually was.

[eminence]

I don't have Wilt as valuing good coaching, he voluntarily left Hannum to go make movies.

I think that’s right to some degree. Wilt also did get almost twice as much money from the Lakers, though, and the opportunity to do things in LA that would make him more money. And that surely mattered quite a lot in a context where the money Wilt was making, even after almost doubling his salary at LA, was less than the veteran’s minimum now (in inflation-adjusted terms). Playing basketball was not the kind of money-making venture it is now, such that someone couldn’t really afford to leave other financial opportunities on the table. Not that this really matters for purposes of how good a basketball player he was, except just that I think his willingness to leave a good coach was reasonable given the financial context of the NBA back then.

Wilt demanded out of Philly prior the end of the '67 season and well prior to receiving his contract offers from Seattle and LA over a year later (Seattle made the initial big money offer which LA then matched and he preferred LA).

[One_and_Done]

Honestly the negative Wilt stories are epic, and will need to get more discussion in the next few threads. This is a guy who refused to live in Philly, so he commuted from NY. He was faking an injury during a contract dispute, then got so fed up with how his team was playing he came and checked in from the stands. Wilt was the Shaq of his time, and not always in a good way.

[Moonbeam]

Thanks - I like the line graphs a lot better.
Interesting on the offensive side in that for most of these guys we don't see that much of a plus or minus , and I think some is attributed to the aging of Arizin (negative for Wilt) and Kobe (positive for Shaq).

Is it saying that there isn't THAT much of an offensive increase in teammates based on these guys? Wondering how other elite playmakers or gravity guys - like Nash/Harden and Curry - compare?

Here are those 3 players:







These values are merely correlations between the change in teammate OWS/48 or DWS/48 and the associated change in a player's MP with those teammates. I'd say it's not the same thing as pure "lift" for teammates, but merely a reflection of whether the players in question tend to be overcredited or undercredited by Win Shares in evaluating a team's performance on offense and defense, respectively.

But the Magic thing on D is really fascinating. Here's a snapshot of 1988, for example:

Code: Select all

Player               MP   DWS  1-yr DCor 3-yr DCor 5-yr DCor
A.C. Green          2636 3.325   0.440     0.586     0.579
Billy Thompson        38 0.045   0.404     0.276     0.165
Byron Scott         3048 3.228   0.425     0.506     0.515
James Worthy        2655 2.323   0.519     0.728     0.627
Jeff Lamp              7 0.001   0.431     0.360     0.475
Kareem Abdul-Jabbar 2308 2.673   0.387     0.445     0.368
Kurt Rambis          845 1.262  -0.383    -0.652    -0.526
Magic Johnson       2637 3.048   0.387     0.709     0.614
Michael Cooper      1793 1.769   0.339     0.714     0.191
Mike Smrek           421 0.583  -0.463     0.083    -0.330
Milt Wagner          380 0.225  -0.306    -0.306    -0.306
Mychal Thompson     2007 2.209   0.461     0.127     0.037
Ray Tolbert           82 0.103  -0.305     0.642     0.642
Tony Campbell        242 0.243   0.434    -0.094    -0.236
Wes Matthews         706 0.573  -0.416     0.073     0.106


I'd have expected Cooper and Kareem to have the strongest set of numbers across the 1-yr, 3-yr, and 5-yr correlations, and while they are positive, they aren't as consistently strong as the likes of Magic, Worthy, and Scott.

[Moonbeam]

There's a suggestion here that Wilt is overcredited by OWS but undercredited by DWS.

Great post, Moonbeam.

I wonder, what would Wilt's DWS look like with blocks and steals recorded? How much of a difference would that make?

It's tantalizing to think about how our perceptions of older players could be more informed by stats like that (or +/- stats). I suspect Wilt's raw DWS would climb at the expense of teammates if we had blocks and steals for those seasons, to the point he would no longer appear undercredited and may even appear overcredited.

[Moonbeam]

Very interesting, especially Magic’s. There’s a distinct possibility that he has been underrated on defense, given that his offense was so good thereby limiting transition opportunities in an era in which transition defense was nowhere near as good given how many players used to crash the offensive boards.

His defense itself doesn’t really stand out when you watch him play as he doesn’t have a particularly high motor, doesn’t aggressively contest shots, doesn’t disrupt many actions, though he does defend the post decently well, which of course was important in his day. Part of his defensive reputation also comes from the comparison to his contemporary, Larry Bird, who had a nonstop motor, was making weakside steals, weakside blocks, using those quick hands, crashing the defensive boards for contested rebounds after being out on the perimeter.

Magic seems highly aware of the offensive actions that are being run—in just about any clip you’ll see him defending at the top of the key, arms out, looking left and right to see what actions are being run, but not necessarily ball watching since in the 1980s you’re not going to have somebody pull up from 30 feet or run a high screen roll out there.

He looks great in Squared’s partial RAPM, a big plus defender in 1985, the a lot of that could be the Kareem still had great defensive impact all the way up until 1985. that team had heavy minutes played by poor/bad defenders in James worthy and Byron Scott, so not sure how much of it was Magic helping them be the 7th ranked defense, -.9 rDRtg.

Have you run any type of regressions to see indications of correlations between MP and delta win shares being correlated with known RAPM data?

Regressing these correlations against RAPM data would be an interesting approach, but I think some of these correlations are perhaps too highly correlated themselves with rosters that have little turnover (as in the 80s Lakers). It's still an idea worth exploring.

[Moonbeam]

Just so I make sure I understand what’s going on here, and what it means, I have a foundational question:

Isn’t this essentially ultimately just getting at a question of how these players’ teams’ changes in win totals year-to-year changed depending on how their minutes-played changed?

Like, ultimately, a team’s total number of win shares definitionally equals the team’s number of wins. So, if your team tends to win a lot more games when you play more minutes, then that extra winning will definitionally manifest itself through some combination of your own win shares going up and/or your teammates’ win shares going up. This measure feels like it is getting at the question of where that balance is for certain players—i.e. whether the amount you increase your team’s wins by is more or less captured by increases in your own win share totals than it is for other players. That’s interesting info. But it feels to me like the way we’d ultimately use that info would just be to layer on that teammates’-increased-win-shares effect on top of the player’s actual win shares, to try to more accurately measure the player’s effect on the team. However, if we’re doing that, given that team win shares definitionally equal team wins, couldn’t we just get to the exact same place by running a regression that charts out the effect of changes in a player’s minutes with changes in the team’s wins? It feels to me like this is a more complicated way of getting to that.

You've nailed the approach I'm taking and how I've been interpreting this data. I'll note that Win Shares aren't definitionally equal to a team's actual wins and losses, but rather their expected wins or losses as functions of team ORating and DRating relative to league average. Nonetheless, you are right in that I'm essentially trying to drill down to see how a change in a player's MP on a team (which is particularly notable when it changes a lot, especially if a player changes teams) impacts team performance. Indeed, those changes will be reflected by a change in the player's own assigned Win Shares as well as those of his teammates. My goal is to try to combine these into "adjusted" Win Shares and use this as a foundation of player evaluation by modelling regular season and postseason performance. I've done this previously using aging curves to come up with expected win shares and had started to write things up in this incomplete draft, and I may flesh this out with an approach using these correlations instead.

The TLDR of it is basically that I'm aiming to come up with an adjusted Win Shares for a player that is equal to assigned Win Shares + delta, where delta is some function of teammate performance relative to expectations.

The reason I don't think I can simply model a change in team performance vs change in a player's MP is that we'd be stripping away any (imperfect) measure of how much a player contributes to these changes. Imagine a scenario where Player A (a great defender) and Player B (a weak defender) have high correlations in their MP for the same team over a number of seasons, and Player B plays half the minutes of Player A. It would be hard to distinguish fluctuations in team performance as affected by Player A vs Player B. In fact, with fewer minutes played, if the team's defense tended to be good, Player B would likely have a stronger signal than Player A.

[Doctor MJ]

Going to drop another post right now pertaining to Wilt, that I'd like to see people chime in on. Before I get to my thing, I want to compliment the points made pertaining to Wilt already, and I'll single out ZeppelinPage in particular for giving patient tutoring on the Dipper:

Spoiler:

For me, I just keep going back to a fundamental basketball thing:

If you're taking what the defense will give you, then you shouldn't have a shoot-mode vs an assist-mode when you have the ball that takes place over extended periods of time like this. You make the best play. If they put to much individual pressure on you, exploit that with a pass, when they ease up, exploit it with scoring.

As 70sFan mentioned earlier I just don't think there is enough evidence of Wilt not doing this, especially when on film he is hitting cutters and passing off doubles just like Kareem would do. I mean, he was averaging 5 assists in '64 and '66.

I understand that Wilt was told to go for maximum points in '61-62, but he wasn't being told to go for maximum points in, say, '65-66, so why is it in '66-67 he dropped from being the guy who shot more FGA than anyone else on the team per minute to 9th out of 11 the next? And why did it work so, so much better? (Note, by TSA Wilt's a bit higher than 9th and I can calculate that if requested.)

Regardless of what we conclude for the RealGM 100, to me those questions are the most important thing for people to be asking and trying to understand.

Wilt isn't going for maximum points in '66 anyways. He's leading the entire team in assists! Factoring in offensive rebounds and tip-ins from the pace he's shooting around the same amount as Hakeem, Shaq, or any other all-time center is while averaging 5 assists. On film he's also passing and hitting players while still scoring similarly to how Kareem would.

Back to you Zep: I'm not alleging that these same limitations are on you. I know you care more about history than Seth does.

I appreciate that; right back at you. My main agreement with Seth, from a broad perspective on NBA history, is that we both see immense value for a team when a player consistently leads the league in both FGA and FG%. So we tend to scrutinize the team and coaches surrounding the player in question, rather than placing blame on him. I don't believe that passing to players like Guy Rodgers and Woody Sauldsberry, who have the two lowest career totals in TS Added, would have resulted in a better offense. I think a major difference in 1967 was the coaching change that instituted a new offense and allowed for improvement in guys like Chet Walker, Wali Jones, and Billy Cunningham to be open more often.

If Wilt were capable of achieving a normal balance between shooting & passing that was optimal in his environment based on what the defense gave him, why the drastic swings? I don't think it's realistic to think that when Hannum comes in in 1996 he insisted that Wilt shoot way less than everyone else. I think he gave Wilt a new focus, and this is what happened. A player already recognizing passing opportunities and making them as a matter of course wouldn't have seen such a drastic swing simply, but with Wilt it did, and that swing would go far more extreme in the years to come.

Was it was best for the team? Quite possibly...but if Wilt were good enough at making shooting vs passing decisions, why would it make sense to do this? It's not like he couldn't do that stuff while also getting rebounds and being a defensive threat.

I hear you and understand this viewpoint. A fundamental difference between our views is that I don't believe Hannum joined the team and specifically changed the offense because of Wilt. I think the modifications were more for the benefit of his teammates than anything. Under Schayes, there were fewer opportunities for Wilt to pass (despite being a willing passer). However, when Hannum implemented his "wheel offense"—a strategy involving players circulating around Wilt throughout a possession, it provided more openings for Wilt, which increased his opportunities to pass. With more talented teammates, this kind of offense worked wonders along with Wilt scoring so efficiently. But I think whether Wilt was scoring or passing more, the '67 team was going to function well around him like this. I blame the offensive results previous to 1967 more on Wilt's teammates and coaches than himself. Hannum had tried to institute this same wheel system with the Warriors in 1964 but the team had zero offensive talent and it actually made the offense worse compared to '62 and '63 because poor offensive players were shooting.

One thing I have to say is that I have never understood this scoring mode/passing mode thing that Ben Taylor came up with based on limited footage.

In 1967, Wilt is clearly not shooting much; he's fully engaged in "passing mode", not integrating his scoring and passing abilities. Yet, they have the greatest offense ever that year with him playing this way. He's not blending his scoring and passing, an approach considered crucial—yet he anchors this all-time great offense?

In December 1968, he reverts to his 1965 persona and shifts into "scoring mode", averaging 30/5 (nearly exactly what he did prior to '67), leading to a 15-2 record. He then reverts to full-on "passing mode" to finish the season and they finish 24-5.

In 1962, he's very clearly in his "scoring-mode" yet their offenses are better than in 1964, 1965, 1966, and not that far off from 1968, yet he was passing more in those seasons than in 1962.

I understand where you're coming from but I just don't buy that the lack of offensive results were his fault when most of his teammates are shooting below league average for their careers before 1967. He's a very efficient high-volume scorer and surrounded with more defensively-oriented teammates until 1965. After that point, his teammates weren't the kind of scorers yet (Chet and Cunningham would eventually improve) and the coaching strategies were lacking. Yes, they improve greatly in 1967 as a result of Wilt in Hannum's offense, but guys like Chet Walker and Wali Jones regress for much of the season in 1968 and the team gets worse with Wilt playing the same way as in 1967. Therefore, I don't think it's as simple as placing the blame on Wilt for the performances of his entire team.

Regardless, I don't believe that the footage or any other sources support this idea that he was not "blending" his abilities in 1964-1966. On film he passes out of double-teams, connects with cutters, and takes scoring opportunities when no one else is open. This is why I believe that the effectiveness of his play hinged more on his teammates and overall coaching than on any aspect of his personal performance.

It's had an effect on me, as has my own further analysis, and Wilt has risen up compared to where I had him in recent iterations of this project.

I do still have find myself bothered by the combination of Wilt's teams not being amazing on offense both during the volume scoring years, and immediately after his team's breakthrough season in '66-67 - along with the fact there was a similar thing that happened on the Lakers.

One would think that if things were as simple as just getting a competent coach who figured out the right way to use Wilt, we should have seen similar team offensive peaks be reached again and again after that, and we just never see such a trend.

We've talked in the past about the decoy theory of Wilt on the 76ers. The idea that the '66-67 offense worked so well because defenses were selling out against Wilt's scoring too much, and when they stopped, a more balanced equilibrium was reached. I don't have the posts handy, but I'll say that people have shown me evidence from splits that would seem to go against this theory, but I did come across something I thought was worth sharing.

There's a site I love called From Way Downtown, which posts old articles that don't exist elsewhere on the internet. This is from an article titled "Wilt Chamberlain: The ‘Shape’ of Things to Come, 1967, published In November 1967.

(Note that if you click on it, you'll see the early part in italics that comes from the bloggers. Don't confuse that from the original piece which is what I'll be quoting from.)

Alex Hannum leaned forward on the bench, cupped his hands, and yelled out, “Shape, Shape,” in a sharp, sure voice. Wilt Chamberlain quickly moved to the high post, just above the free-throw line, as Hal Greer dribbled across midcourt.

The “shape” play was a routine controlled-offensive maneuver, one the 76ers had used successfully hundreds of times before. A simple play based on fear. Just throw the ball into Wilt, and when the opposition folds back frantically to help stop the world’s greatest scorer, you either cut to the basket or move to an open spot for an automatic 10-12-foot popper.

...

Using this kind of strategy, with the muscle and talent to make it go, Hannum’s hotshots last season had gone through the league faster than a wife goes through a paycheck. They wiped out the Boston dynasty, won the NBA title without being pressed to the limit, and apparently started a dynasty of their own.

So "the shape play" is another terminology for what ZeppelinPage referred to as "the wheel", and what I'd say is a descendent of "the pivot play" which originated in the 1920s by Dutch Dehnert and the Original Celtics, all of which are antecedents to what Denver does today with Jokic.

As they say, and as we know, the 76ers used it to create the best season in history to that point in '66-67.

Greer lofted the ball into Wilt. But instead of calling for help and making sure he stayed between Chamberlain and the basket, Bill Russell quickly moved around the 7-foot-3 center, knocked the ball away, and two seconds later Sam Jones was sinking a 16-foot jumper.

This cut the heart out of the 76ers, as plain as if a knife had been used. They fumbled and fouled their way into a 116-111 defeat, making only one of their last 11 shots from the floor and looking like a bunch of unsure amateurs in the process.

Today, they’re in second place in the Eastern Division with a 12-4 record, and Boston is in first at 12-3. This time last year, the 76ers were 15-1, and never looked back after that. Why the big change?

I'm including the last part here to make sure we all raise our eyebrows at this a bit, and note that we need to take it with a grain of salt. 15-1 is an unrealistic thing to expect teams to start with every year no matter how dynastic they are, 12-4 is certainly what a contender looks like, and we know looking back from the future that the 76ers end up getting the #1 seed comfortably.

With that acknowledged, the following quote is what I really wanted people to chew on and speak to:

It’s obvious, so obvious that even Hannum doesn’t try to talk around it anymore. The opposition isn’t afraid of Wilt Chamberlain as a scorer anymore. The big man, who once scored 100 points in one game and averaged over 50 for a full season, all of a sudden has become a so-so offensive player.

“I’m sure the rest of the league has become aware of the fact that Wilt’s having trouble with his offensive moves, that he’s not thinking about the hoop as much as he should,” a saddened Hannum said after Saturday night’s loss at the Spectrum before a record 15,239 mourners.

“They’re not playing him as honest as before, and if you can’t keep them honest on defense, then you’re in trouble. It’s obvious that Wilt has to make some re-evaluations of his offensive game.”

To me this is essentially what's being put forward with the "decoy" model, even as it's clear that "decoy" isn't how the 76ers were thinking about it. What's real here is the gravitational effect of Wilt based on the type of threat he poses.

When Hannum says "they're not playing him as honest", what he literally means is that they are playing Wilt expecting him to pass in a way they weren't the previous season, and in doing so they were getting away without being burnt by Wilt's scoring attack in a way theoretically shouldn't have been able to.

I'll end the meat of the post there except to say that there's a slant to this article that I alluded to with the 12-4 start but exists beyond it in parts I haven't included here. You can certainly bring the slant more to light if you feel it necessary, but what I'm interested in here more than anything else are people's thoughts pertaining to the actual Hannum quote.

Zep I'm not responding you point by point in the post I spoiler-quote above because on the whole I just appreciate your analysis, but I did want to hit one thing specifically:

One thing I have to say is that I have never understood this scoring mode/passing mode thing that Ben Taylor came up with based on limited footage.

So, in terms of scoring/passing mode, that is what I'm talking about above to be clear, but I'm responding here because of the Ben Taylor reference.

It's become quite the thing since Ben hit the big time to attribute people on the PC board specifically as people just "buying whatever Ben Taylor is selling". This is a weirder thing for me than probably anyone else on the board because a) I was running projects on this board years before Ben arrived, b) I ran the project that brough Ben, and other great posters onto the board (RPOY), c) I started my blog just before he started his and we referenced each other like crazy at the time, and d) he and I are very good friends who have been having deep basketball conversations regularly ever since.

I'll frankly say flat out: It's not random that he's the one who made it big. To use baseball parlance: Ben's a 5-tool player at this.

I am not jealous of Ben's success in the slightest - he deserves everything he's got and I really couldn't do what he does - but I do get irked at the idea that the PC board has ever been just a bunch of Ben cult followers. I see it as the place that was the best place on the internet to talk about basketball when I was looking for that place in 2005, and the place I helped grow into a unique scene.

But perhaps more amusingly: If you think Ben sees this board as the place where his followers are, do think again. The difference between the PC board and his Discord/YouTube community is just plain huge, and not in a "they love him even more" way which implies he's literally beloved here.

Over on the GB, someone posted in a thread that because the 2023 Top 100's Top 3 was the same as Ben's Top 3, that was evidence that the latter caused the former. I'd say that the truth is more that PC Board folks who participate in the 100 tend to look at a lot of the same data Ben does. Now at this point some of that data is coming directly from Ben - his site has become a really great resource over time - but people don't use his stats because they think they're perfect, they use them because the concept resonates with how they're looking to analyze the game.

If we try to think about various places that talk about basketball to triangulate PC Board culture, I might say it's a mix of the GB, APBR & APBRmetrics. By that I mean:

GB: Where the Late Stage Social Web mixes with RealGM - the great independent basketball message board where the more thoughtful basketball-internet settled after ESPN bought sportstalk.com and turned it into their Community Forum where it became about as toxic as toxic could be in those times when the internet was still ruled by lolcats.

APBR: The message board for basketball history. Very pure, not a lot of traffic or debate, but great to read.

APBRmetrics: Basketball's answer to SABRmetrics in message board form (courtesy APBR), which became THE place to get noticed if you wanted a team to hire you as a statistician.

The PC Board is...
More GB than the APBR or APBRmetrics.
More APBR than the GB or APBRmetrics.
More APBRmetrics than the GB or APBR.