#4 Highest Peak of All Time (Wilt '67 wins)

[ElGee]

I've said many times the in/out section of my site needs to be updated, mostly because of SRS and schedule and whatnot...that said, just check the 8 magical games you've concocted for Bird here. I'm not seeing an error on my end -- perhaps you could explain how you have the results you have in just the 8 games (seems easy enough)?

[Dr Positivity]

Robertson 64 - Seems to have a terrific combination of scoring, passing, rebounding. Not sure whether to pick this season or another but I'll throw it to his MVP year for now despite getting clowned by the Celtics, he seemed to have performed well

With all due respect, is this spoken from knowledge of the situation, or is this based on a presumption that "since he's Oscar Robertson, one of the greatest players in the history of the game and won his only MVP that season, he had to have performed well?" I actually have the details if anyone's interested.

I went by the boxscore stats which seemed on the level of his regular season. I would appreciate the extra information if you have it if there's any Oscar playoff run that stands out over another to you that deserves the vote over 64 (ie '63)

[lorak]

Constitution gives him right to have such opinion ;]
But the fact remains that clutch is something what matters. Not in a sense, that points scored during last 5 minutes count more, no. But when game is tied at the end, everybody plays at maximum concentration level, defenses are focused as much as possible and that’s why most people value so highly players, who are capable of scoring in such circumstances. And in no way Duncan was as clutch as Hakeem. Never.

You just aren't understanding still. All the stuff that happens at the end of the game is PART OF THE GAME. All the metrics we are using to evaluate players, INCLUDES CLUTCH already. Since no possession at any point in the game can count more (although garbage time could count less), you can't extract more value from your play at the end of the game.

No Elgee, you don't get it. I'm not saying that pts scored during last 5 minutes of close game count more. That's silly, because every point count the same. I'm saying - players and teams are different at the end of close games, more focused, concentrate at 100% level as much as possible. And that's why so many people value ability to score under these circumstances (lets name them - extremally stressful) more that during rest of the game.

[lorak]

Some interesting stats. In 1987, Bird missed 8 games in a season right after his prime.

8 games Bird missed:
5-3 W-L (4.57 SRS)
115.1 O rating vs 108.3 opp Defense (+6.77)
109.4 D rating vs. 107.3 opp O rating (+2.16)

So the offense actually played better than when Bird played but the defense was also worse than when Bird played.


We keep discussing portability, but we also have to talk about Defensive portability. LeBron has proven he can play elite defense on many kinds of defensive teams like one with a big man anchor (09 Wallace), one that is perimeter based (Miami), and defenses with small guys in other positions (Mo, Delonte, Joe Smith, Varejao). How would Bird do defensively if he had weak big men behind to clean up his mess? What if he didn't have McHale who would sometimes guard the good opposing SF like McHale did vs. Wilkins.

This is not the information I have at all. First of all, when we have seasons with healthy samples, the 8-gamer is subject to huge variance. That said,

1987 Celtics
w/ Bird (74g) +5.9 ORtg -0.9 DRtg 6.9 SRS
w/out Bird (8g) +5.2 ORtg +0.7 DRtg 3.8 SRS

No notable lineup changes for opponents. Boston was healthy except for Ainge missing one game. The 74g sample with Bird includes 10g w/out Ainge and 5g w/out McHale. Again, we see no difference if we drop the 5g w/out McHale -- +6.7 SRS, +5.6 ORtg. (In the 5g McHale missed, Boston had a +9.2 ORtg and a 9.4 SRS). It's possible there's I have an error somewhere, but I didn't see one at quick glance...

it's possible your numbers are wrong. For example on your site in the in/out section, you had the Cavs at +4.1 in the games LeBron played from 08-10, but thats not possible because I have the Cavs outscoring opponents by 1238 points (5.81 PPG) from 08-10 in the games LeBron played.

As far as that 8 game sample, I have the celtics outscoring opponents 893-849 with 775.8 possessions played in 8 games. So thats a +5.5 MOV with a -0.93 SOS which gave me the 4.57 SRS

According to my numbers (based on basketball-reference game log) 1987 Celtics:
with Bird 113.8 ortg, 107 drtg
without Bird 113.6 ortg, 108 drtg

[ElGee]

I have had this discussion a dozen times. Trust me for a moment and consider the possibility that you are not understanding this. Can we consider that possibility?

Excellent. People have weighting problems in their brains. They can't figure out large samples. They put too much stock in what they see recently. There is no special reward for being better in stressful situations, since it's part of who you are overall as a player. Your last sentence is a logical error. I'm not sure if you've read the discussions and articles I've posted about this but it's been covered, you are just saying "I'm not talking about points count more...I'm talking about possessions counting more!"

[colts18]

I've said many times the in/out section of my site needs to be updated, mostly because of SRS and schedule and whatnot...that said, just check the 8 magical games you've concocted for Bird here. I'm not seeing an error on my end -- perhaps you could explain how you have the results you have in just the 8 games (seems easy enough)?

I got these 8 games, I got the point scored/allowed and used the pace numbers from the box score:

http://www.basketball-reference.com/box ... 60BOS.html
http://www.basketball-reference.com/box ... 80DET.html
http://www.basketball-reference.com/box ... 70BOS.html
http://www.basketball-reference.com/box ... 90BOS.html
http://www.basketball-reference.com/box ... 00DET.html
http://www.basketball-reference.com/box ... 30BOS.html
http://www.basketball-reference.com/box ... 50BOS.html
http://www.basketball-reference.com/box ... 60CLE.html

For that, I got 893-849 (+5.5). Then for SOS, I got:
Clipper -11
Pistons 3.52
Pistons 3.52
Bucks 4.04
Kings -3.33
Nuggets -1.14
Sixers 0.12
Cavs -3.18

That adds up to -7.45 SOS plus 44 MOV which equals 36.55. 36.55 divided by 8 is 4.57 SRS

[lorak]

I have had this discussion a dozen times. Trust me for a moment and consider the possibility that you are not understanding this. Can we consider that possibility?

Excellent. People have weighting problems in their brains. They can't figure out large samples. They put too much stock in what they see recently. There is no special reward for being better in stressful situations, since it's part of who you are overall as a player. Your last sentence is a logical error. I'm not sure if you've read the discussions and articles I've posted about this but it's been covered, you are just saying "I'm not talking about points count more...I'm talking about possessions counting more!"

Tell me, do you really think that players play at maximum concentration level during whole game?

You are just wrong because you look at "clutch problem" from wrong angle ("pts scored at the end count more"). While the thing is: late close game situations are the situations when players play at level closest to maximum concentration. That's how human mind works. Because of that, these conditions are the hardest during the game and thus players, who produce the most under these conditions are the best, because they give the highest probability of winning (because they are the best under the most difficult circumstances). No logical error in that, and winning probability is the most important thing in games (so also in basketball).

[ElGee]

@ Colts -- you are doing two things wrong: you can't average the ORtg across games because the # of possessions is different in each game (Stern's numbers are correct here). Second, you don't ADD SRS. You need to average the opp SRS to achieve SOS, which yields a -0.93. (-7.45 SOS would be the easiest stretch of games in NBA history.)

@Stern -- ALL I care about is winning probability. I've done extensive research on winning probability at the end of close games. Have you read it?? You will find that the best clutch players in the world can probably add 2 wins a year to a team, which is less than an SRS point, which is already part of the overall analysis of the team/player. It's reflected in their points scored/allowed ALREADY! In the PS, such a super clutch player would, on average, add 0.5 wins for a full PS run, or every TEN playoff series he might be worth an extra win in a game an average clutch player would lose.

[Doctor MJ]

Writing in blue bold just to make sure people see this. A few things:

1. To anyone uncertain about rules in this project, see the first post of the Project thread or PM me.

2. Specifically, people seem confused about how I tally up votes. I don't carry votes over across threads. If you still are set on the same guy as last thread, you've got to say it in the current thread, ideally while re-posting your previous logic.

3. There is a Official voting panel (go to the Project thread to see it). If you aren't on it, don't vote. If you do vote, it won't count anyway, and I'll be annoyed with you. If you want on the panel, PM me for consideration, and then participate in the discussion so I know you mean business.

Finally...

Conversation seems a little dicey. Some people are getting irritated with each other. In thread #3 several people didn't vote. In general, people are getting frustrated because it feels like there's a lot of talking passed one another rather than real, informative debate.

Please try to be patient with each other, and to be diligent about really reading other people's posts.If you get busy for a few days, 'salright, we all know that happens, but people shouldn't forget that the actual rankings are just the dessert.

The main course is the discourse. We're here to learn.

Last, if you realize you're really going to be away for a few threads, please PM to let me know. This isn't something for your to worry about because you go to Vegas (or whatever your local place of debauchery is) for the weekend, but if something really comes up, just let me know.

Cheers,
Doc

[JordansBulls]

From what I'm seeing so far:

09 LeBron - 4 (Doctor MJ, colts18, DavidStern, therealbig3)

67 Wilt - 3 (ardee, PTB Fan, Dr Positivity)

94 Hakeem - 1 (C-izMe)

71 Kareem - 1 (Josephpaul)

I took Kareem 1971 as well. But I just want to know why not more love for Kareem 1971? I just don't get the Lebron 2009 especially losing to a team that he was favored over. I can understand maybe if he lost to the defending champs in the Celtics with KG around or LA but losing to Orlando who was missing it's starting PG.

Once again, because '71 Kareem wasn't even Kareem's own best season. If someone has a full career, it's simply impossible for them to peak in their sophomore season.

Let me put it to you this way. You're the biggest Jordan fan on the board. Could you in ANYWAY justify someone picking '87 Jordan (his second full season) as his best year?

Doesnt apply because MJ 1991 had better stats and more success than 1987 MJ. In this case Kareem in 1971 and 1972 had better numbers than Kareem 1976 and 1977 and less playoff failures on top of that. At least in 1971 he won it all and in 1972 he lost to clearly superior teams. In 1976 that doesn't help his case and in 1977 he lost with HCA in a sweep.

[colts18]

@ Colts -- you are doing two things wrong: you can't average the ORtg across games because the # of possessions is different in each game (Stern's numbers are correct here). Second, you don't ADD SRS. You need to average the opp SRS to achieve SOS, which yields a -0.93. (-7.45 SOS would be the easiest stretch of games in NBA history.)

I didn't average the O ratings. I added up all the possessions from each game based on what was said on the B-R boxscore. That gave me 775.8 possessions in 8 games. Then I did 893/775.8*100 which got me the 115.1 O rating.


No, Im pretty sure I did the SRS part right. SRS is MOV + SOS. The MOV was 44 points. The SOS was -7.45. That adds up to 36.55 then divide that by 8 games which gets you 4.57 SRS

[Doctor MJ]

You are just wrong because you look at "clutch problem" from wrong angle ("pts scored at the end count more"). While the thing is: late close game situations are the situations when players play at level closest to maximum concentration. That's how human mind works. Because of that, these conditions are the hardest during the game and thus players, who produce the most under these conditions are the best, because they give the highest probability of winning (because they are the best under the most difficult circumstances). No logical error in that, and winning probability is the most important thing in games (so also in basketball).

DS, just jumping in here, so forgive me if there's something I missed:

The key assumption I think a lot of people make is that because, as you say, players are most focused in crunch time, the better crunch time players are able to provide a bit of extra oomph when it matters to put their team over the top, and that this oomph will not be able to be simply seen as a part of the overall game because the oomph is something will vary quite a lot depending on how badly it's needed.

However, there are still ways to measure the oomph. The goal of the oomph is to make your team win close games, so a team has a serious oomph edge, you'd expect them to consistently have a better W-L than SRS says they should.

The reason why the anti-clutch geeks exist is because they've already gotten as far as analyzing that data, and what they find is that the players who everyone agrees are "clutch" typically aren't showing any real trend of this disproportionate winning of close games.

On the other hand, they are seeing that "clutch" players are easily identified statistically simply based on how much the monopolize their team's shots in the clutch, which isn't necessarily a good thing at all. So the entire basis behind a typical person's clutch fetish is quite clearly misguided.

This doesn't mean that there aren't better and worse performers in the clutch of course, but it calls into question the scale of their relative "oomph" quite severely.

[ElGee]

@ Colts -- you are doing two things wrong: you can't average the ORtg across games because the # of possessions is different in each game (Stern's numbers are correct here). Second, you don't ADD SRS. You need to average the opp SRS to achieve SOS, which yields a -0.93. (-7.45 SOS would be the easiest stretch of games in NBA history.)

I didn't average the O ratings. I added up all the possessions from each game based on what was said on the B-R boxscore. That gave me 775.8 possessions in 8 games. Then I did 893/775.8*100 which got me the 115.1 O rating.


No, Im pretty sure I did the SRS part right. SRS is MOV + SOS. The MOV was 44 points. The SOS was -7.45. That adds up to 36.55 then divide that by 8 games which gets you 4.57 SRS

Look, this will be the last time I address this, and it's so people reading this project just know what the right numbers are. And the numbers you have posted aren't right, and there are reasons for this.

Boston played an overtime. The pace you tried to average, it adjusts for overtime. So in the OT game, you get the right number of points but not the right number possessions. Make sense?

Second, you are forgetting to account for how many games they play in SOS calculations. The Celtics SOS in the games you posted wasn't close to -7.45, because you AVERAGE the opponent SRS's. You need to take that -7.45 you have added up, and divide it by 8 games played (-0.93). That's the Opp SRS. And then you should add a HCA factor, which typically is 3 points (although it was closer to 4+ on average in the 80s).

Cheers.

[colts18]

Look, this will be the last time I address this, and it's so people reading this project just know what the right numbers are. And the numbers you have posted aren't right, and there are reasons for this.

Boston played an overtime. The pace you tried to average, it adjusts for overtime. So in the OT game, you get the right number of points but not the right number possessions. Make sense?

Second, you are forgetting to account for how many games they play in SOS calculations. The Celtics SOS in the games you posted wasn't close to -7.45, because you AVERAGE the opponent SRS's. You need to take that -7.45 you have added up, and divide it by 8 games played (-0.93). That's the Opp SRS. And then you should add a HCA factor, which typically is 3 points (although it was closer to 4+ on average in the 80s).

Cheers.

I see on the OT Part, i'm not sure why B-R doesn't put the real pace on the box score. Though I guess we have a different philosophy on the SRS because I don't add HCA since HCA is not included in the usual SRS calculations.

[ElGee]

Getting back to business...I'm really feeling like people aren't feeling Bird here. Meanwhile, I'm having a hard time seeing an argument for peak Magic over peak Bird, because I really can't see a clear separation in their offense. Someone pointed out their defense/rebounding similarities, but at their peaks I don't entirely agree with that -- Magic looks like average or slightly below average on that front, strong arguments can be seen for Bird being a positive (and two years removed from respectable all-D honors).

My guess is that people who are leaning LeBron see his offense close to Bird's that James' incredible defenses make it a trump card. To that end, I haven't seen a compelling argument, because I don't find it compelling to play a unipolar role on a bad offensive team and get them to a respectable level. The argument has been made that the 09 Cavs hit much higher than respectable levels, but I hope my last few posts on the matter have explained why I disagree:

(1) Those 3-point shooters were shooting at a historically good rate...they deserve some of that credit.

(2) The ensuing season, we did not see such a level of offense with very minor tweaks. From where I stand (obsessively watching analyzing games in 2010), James wasn't really "worse" (if anything, I side with those saying he was better). So this makes it strange to think of James' QBing act as leading incredible offenses, unless you happen to believe 09 James was an aberration.

I've presented data from 3 different versions of Bird and 3 different team settings (early 80s w/Tiny and Parish, peak w/DJ, 2 jump-shooters, McHale post and Parish mid-range, past prime Bird on super-balanced teams). Bird's offensive lift looks incredible, and this jibes with the way he plays. I'm not sure if everyone here has seen peak Bird play -- there are full games now from the 86 or 87 season up on youtube. Everyone should watch one or two (or even highlights) before voting here if they don't have a feel for the difference between LBJ and Bird. Bird is a solid 6-9 -- he gets his shot off over basically everyone. He's a GOAT-level shooter, a GOAT-level off-ball player, a GOAT-level passer...I mean, the old "eye-test" is lining up with numbers we have available.

Let me use a numerical illustration to again hammer this home about unipolar performances. The first number is a team's ORTG w/out the player, the second with the player:

Player A ORTg change
-5 to 3
-3 to 4
0 to 5
+3 to 6
+5 to 7
+8 to +10

Player B ORtg change
-5 to -1
-3 to 1
0 to 4
3 to 7
5 to 9
8 to 12

Assuming their defense is equal, who do you think helps win more championships?

Everyone wants to say Player A. That should be your default instinct. I'm not going to make a separate thread and have people guess and whatnot, I'm just going to tell you Player A is not the right answer.

Player A Championship Odds: ~14%
Player B Championship Odds: ~18%

Why is this so relevant? Because the tendency is to look at 94 Hakeem, 03 Duncan, 09 LeBron and any other unipolar act and glorify the living hell out of it. But if that player has stronger diminishing returns on better teams, that player won't win you as many championships. Notice in the hypothetical I drafted here that Player A would be an MVP front-runner on his weak teams and Player B would look like a mere all-star on weak teams. But the ability to help average and good offenses is simply SO much more important than helping bad ones.

This doesn't mean that guys who help bad offenses can't scale to good ones -- you need to "guess" this yourself if he's never put in a different situation. It just means people should be very, very very careful about overrating the effect of some player "carrying" teams, especially when the heights they carry them to in the first place aren't all that high.

[ronnymac2]

I never said that team defensive rankings prove Duncan was better than Hakeem. I even said that Hakeem could very well have been a better defender.

I am pointing to the fact that Duncan anchored better defenses, and I'm asking the question why? I know he had Popovich and good defensive role players, and I asked, "How much credit should we give them"? Because even with those players around him, and even with that coach around him, the Spurs defense has fallen dramatically after 07, when Duncan gradually fell out of his prime. So the elite Spurs defense seems to depend pretty strongly on Duncan's ability to anchor it.

I am also pointing to the fact that Duncan was consistently among the best in the league in terms of defensive RAPM, and was probably the best defender in the league during his prime.

Hakeem missed 26 games in 91. His team posted an average 105.8 DRating vs an average ORating of 108.2 (-2.4). On the year, they were -4.0 defensively. Rough math, but with Hakeem, it seems they were around -4.7. So they go from a -4.7 defense with Hakeem to a -2.4 defense without Hakeem. Decent lift, but nowhere near catastrophic without him, and it doesn't seem to be near the kind of impact that he's described as having. And isn't this earlier, freakishly athletic Hakeem, who was supposed to be better than 94/95 Hakeem defensively?

Duncan missed 16 games in 05. His team posted an average 105.1 DRating vs an average ORating of 106.4 (-1.3). On the year, they were -7.3 defensively. Again, rough math, but with Duncan, it seems they were around -8.8. So they go from a -8.8 defense with Duncan to a -1.3 defense without Duncan. That seems like pretty huge lift, and is much better than what we saw from Hakeem in 91. And it seems that Duncan is the difference between that defense being average and that defense being elite, at least in 05.

I know 05 Duncan vs 91 Hakeem isn't being discussed here, but 91 Hakeem was supposed to be superior defensively than 94/95 Hakeem, and 05 Duncan was supposed to be inferior defensively to 02/03 Duncan...and Duncan still seems to have superior defensive lift. So I'll ask again:

Other than looking prettier defensively, where is the evidence that Hakeem was actually superior?

Also, it would be interesting to check out how the Rockets and Spurs did offensively without Hakeem or Duncan.

OK, so Doc MJ's responses were similar to what mine would have been. I won't go too in depth because there's been an enormous amount of other cool subjects covered since yesterday.

Duncan's defensive decline coinciding with SA's defensive decline could be the shift in strategy from a focus on defense to a focus on offense that Coach Popovich has been lauded for over the past bunch of seasons. Ginobili and Parker have had way more control, and San Antonio has went from pairing Duncan with another interior-oriented big man (Robinson to Rose to Nesterovic to Elson) to basically surrounding him with four smalls or pairing him with Matt Bonner or Boris Diaw.

It's a possibility. A strong possibility.

As far as Duncan measuring out as the best defender of his era- sure, I could agree with that. Hakeem didn't play in Duncan's era though, so I'm not sure how helpful that finding is for us.

What did Hakeem do better defensively? Again, one of our doctors (we have so many in these projects!) said what I would have said.

[Dr Positivity]

Bruce Bowen also left a vacuum defensively for the Spurs on the perimeter

[ronnymac2]

^^^Absolutely. And instead of finding a suitable replacement defensively, they found wings that could shoot 3's like Bruce but had a few more offensive tricks at their disposal.

[mysticbb]

I see on the OT Part, i'm not sure why B-R doesn't put the real pace on the box score. Though I guess we have a different philosophy on the SRS because I don't add HCA since HCA is not included in the usual SRS calculations.

HCA is not part of the normal calculation, because all teams have the same amount of games at home and on the road. But in a selective sample there will be differences in terms of home and road games. Not accounting for that is silly, to say the least.
The Celtics in the 8 games Bird missed played like a 3.77 SRS team, that accounts for HCA and opponents.

[colts18]

Getting back to business...I'm really feeling like people aren't feeling Bird here. Meanwhile, I'm having a hard time seeing an argument for peak Magic over peak Bird, because I really can't see a clear separation in their offense. Someone pointed out their defense/rebounding similarities, but at their peaks I don't entirely agree with that -- Magic looks like average or slightly below average on that front, strong arguments can be seen for Bird being a positive (and two years removed from respectable all-D honors).

My guess is that people who are leaning LeBron see his offense close to Bird's that James' incredible defenses make it a trump card. To that end, I haven't seen a compelling argument, because I don't find it compelling to play a unipolar role on a bad offensive team and get them to a respectable level. The argument has been made that the 09 Cavs hit much higher than respectable levels, but I hope my last few posts on the matter have explained why I disagree:

(1) Those 3-point shooters were shooting at a historically good rate...they deserve some of that credit.

(2) The ensuing season, we did not see such a level of offense with very minor tweaks. From where I stand (obsessively watching analyzing games in 2010), James wasn't really "worse" (if anything, I side with those saying he was better). So this makes it strange to think of James' QBing act as leading incredible offenses, unless you happen to believe 09 James was an aberration.

I've presented data from 3 different versions of Bird and 3 different team settings (early 80s w/Tiny and Parish, peak w/DJ, 2 jump-shooters, McHale post and Parish mid-range, past prime Bird on super-balanced teams). Bird's offensive lift looks incredible, and this jibes with the way he plays. I'm not sure if everyone here has seen peak Bird play -- there are full games now from the 86 or 87 season up on youtube. Everyone should watch one or two (or even highlights) before voting here if they don't have a feel for the difference between LBJ and Bird. Bird is a solid 6-9 -- he gets his shot off over basically everyone. He's a GOAT-level shooter, a GOAT-level off-ball player, a GOAT-level passer...I mean, the old "eye-test" is lining up with numbers we have available.

Let me use a numerical illustration to again hammer this home about unipolar performances. The first number is a team's ORTG w/out the player, the second with the player:

Player A ORTg change
-5 to 3
-3 to 4
0 to 5
+3 to 6
+5 to 7
+8 to +10

Player B ORtg change
-5 to -1
-3 to 1
0 to 4
3 to 7
5 to 9
8 to 12

Assuming their defense is equal, who do you think helps win more championships?

Everyone wants to say Player A. That should be your default instinct. I'm not going to make a separate thread and have people guess and whatnot, I'm just going to tell you Player A is not the right answer.

Player A Championship Odds: ~14%
Player B Championship Odds: ~18%

Why is this so relevant? Because the tendency is to look at 94 Hakeem, 03 Duncan, 09 LeBron and any other unipolar act and glorify the living hell out of it. But if that player has stronger diminishing returns on better teams, that player won't win you as many championships. Notice in the hypothetical I drafted here that Player A would be an MVP front-runner on his weak teams and Player B would look like a mere all-star on weak teams. But the ability to help average and good offenses is simply SO much more important than helping bad ones.

This doesn't mean that guys who help bad offenses can't scale to good ones -- you need to "guess" this yourself if he's never put in a different situation. It just means people should be very, very very careful about overrating the effect of some player "carrying" teams, especially when the heights they carry them to in the first place aren't all that high.

What about defensive portability? Do you think that exists? I'm not even sure Bird's defense is all that good. For example in 85, I saw him guarding Rambis while McHale guarded Worthy. Then in 87 Mchale still guarded Worthy. And the famous Bird-Wilkins duel was not much of a duel since McHale had to guard Wilkins, not Bird. I'm going to look more into this, but its hard to have Bird high up defensively when he wasn't guarding the best SF of that era.