Greatest Peaks Project (2022): #9 - 1985-86 Larry Bird

[DraymondGold]

...

Excellent post.

I understand you are really high on portability. Are you concerned at all, that by many of the impact metrics we have, that Curry never approached his 16 impact when Durant joined?

RAPTOR, ESPN RPM, BPM, LEBRON, Total Wins (link might be broken) and other stuff consider 16 his best year by far.

Read on Twitter

?s=19

Thanks! Well when the 4 stars are on the court, they're +17, which is a country mile better than the 1996 Bulls, for a 3 year span. To me that's the biggest team outlier stat in NBA history, and it's by a lot. And it includes 2 non-peak years for all 4 of their stars. If there weren't any diminishing returns with that much of an outlier, it would be ridiculous. But like I said in my reply to falcombardi on the last page (https://forums.realgm.com/boards/viewtopic.php?p=100556689#p100556689), Curry shows far less diminishing returns with stars than other top players like LeBron and Wade. So to me that shows portability.

And to be clear, the vast majority of stats I've looked at (including many of those mentioned) have Curry as just as good as he was in 16 at the end of the 17 regular season and through the 17 playoffs. I've argued in past threads (with in-depth stats/film analysis) that Curry's 17 regular season value is lower due to a slump from learning to adjust to Durant (which shows good leadership by making sure KD was comfortable before himself, but of course loses scoring value). Threads are here (https://forums.realgm.com/boards/viewtopic.php?p=100321960#p100321960) and here (https://forums.realgm.com/boards/viewtopic.php?p=100359184#p100359184). Once he figured out how to fit with KD, his value went right back to 16 levels.
__________

On defense: there's an increasing strategy of aggressively picking on defensive liabilities in the playoffs, and more often than not, these players with liabilities are big men who can't guard the perimeter more often than the small player who can't guard the big when there's help defense (the mismatch that generates the open 3 is usually better offense than the mismatch that generates the post-up shot, though again this is a trend and not an absolute rule)
____________
....
Let's take LeBron and Wade as an example. From 2012-2014:
Team SRS without wade: 57-win pace. --> Team SRS with wade: 59-win pace by
Team relative Offense without wade: +8.5. --> Team relative Offense with wade: +7.9.
LeBron's Pts/75 without Wade: ~32.7 --> LeBron's Pts/75 with Wade: ~ 24.5 (by-eye estimate from looking at graph)
LeBron's rTS% without Wade: +8.8%. --> LeBron's rTS% with Wade: +7%

The answer that makes the most sense to me is my interpretation of scalability.
...
Ah, here we come back to the sample size issue. Because Magic actually didn't lead the greatest offensive team of all time, in either 1-year or 2-year samples. Curry's best regular season offense beats any regular season offense Magic ever led, even relative to league. And Curry's best postseason offense beats any postseason offense Magic ever led, even relative to league.

But for peaks, Curry led a greater relative offense than Magic, which I think supports the portability point.

...

When LeBron's fellow stars come on, many of LeBron's individual metrics drop quite a bit. The scoring and efficiency both drop, with a particularly large scoring drop. There's also a drop in wholistic metrics (e.g. APM).
When Curry's fellow stars come on, Curry's individual metrics are far less clear. The scoring drops, but it's over 40% less of a drop than LeBron, and the efficiency actually improves with the smaller volume. There's also less of a clear drop in wholistic metrics.
Hmm... this Curry guy seems to fit better next to good teammates than LeBron's.

When LeBron's fellow stars come on, his team improves by 2 wins. Wait... 2 wins?? That's it?!?
When Curry's fellow stars come on, his team improves from +10.81 (still better than the 86 Boston Celtics) to +17 (significantly% better than the 1996 Bulls!)

So yeah, since Curry's teams are only good when he's on the court, I tend to think Curry's the "special sauce" that makes it work
...
. So true!
...
I think I've explained why I disagree. It's not just theory -- there's plenty of evidence to support it. As to whether you're convinced by the evidence, that's up to you. Let me know if you disagree with anything!

RE: Big men defense

Do you consider perimeter defense for centers a bigger weakness than a perimeter player who gets picked on for being a bad individual defender, is too small or messes the rotations off ball? Cause a lot of offensive perimeter guys have one or many of those limiations (lillard, nash and westbrook come to mind respectively) and teams still played them

A center who is bad on the perimeter but solid in the paint (the more important part of a center job) can be "hid" by playing a scheme that avoids him switching on guards and teams that go with multiple big (like, lets say boston, milwaukee) manage incresible defense by going big even when guys like lopez and horford are not "perimeter centers" un defense by any definition

Great question! I still consider big man on average to be more valuable on defense than perimeter players of course. The point I was specifically arguing for was value today relative to a ~15 years ago relative to ~25 years ago. Big men are far bigger perimeter defensive liabilities today than they were back in the day; meanwhile I wouldn't say perimeter players have lost much value to being liabilities in the post.

If you forced me to pick who's the bigger liability on average today, I am open to the idea that big men are bigger perimeter liabilities in today's era than perimeter players are liabilities in the paint. Why? It personally feels harder to pre-switch the pick and roll or continuous perimeter screens than it is to pre-switch to prevent a big men on smaller defender. Some of this comes down to speed of action -- the faster the action, the harder it is to pre-switch, and it's a lot easier to have quick screening action on the perimeter vs in the paint. The cost of being slower on a screen on the perimeter is also at an all-time high. Meanwhile, help defense is far easier in the paint vs the perimeter. Some examples today might be various star drop bigs in playoff series, like this year's Nuggets vs Warriors, or 21 Jazz vs Clippers, or 19 Jazz vs Houston.

Regardless of which liability is bigger, I still feel big men on average add more value with rim protection than perimeter players do with perimeter defense.

RE:ON-BALL overlap

While it is theorically true, you are taking a case like 2012-2014 heat, a team where halfway through that run wade started to decline with injuries, as evidence

If you look at their offense 2012-2013 before the pkayoffs vs after that, the splits are night and day

You are taking a worse case scenario for on ball overlap here because lebron already can create elite offense with shooters/finishers and without another star so the margin to improve is smaller and then combining him with an athletic slasher/cutter without the athletism (wade post injuries) who is a weaker spot up shooter so he can be ignored off ball and is not a big positive on ball either

If wade played with curry instead i am sure you would see a similar split post injury for wade/curry, is the equivalent of durant playing with a broken hand that doesnt let him jumpshot with the defense knowing it and blaming curry/durant fit for the bad results

Is it really worst case scenario? Sure, Wade declined, but LeBron was at his peak, and I'd have plenty of players as less portable than LeBron. Even if Wade was declining, he wasn't only worth "2 wins" in his older age... that can only be explained with diminishing returns. And even if he's worse, he's certainly still a better player in raw value than whoever they had coming off the bench for him. So why does LeBron's scoring and shooting numbers and the team's offense? The only explanation I can come up with is diminishing returns with better teammates (which is another name for worse fit with better players... which is another name for scalabilty/portability).

More Poor Scalability Examples: And like I mentioned, you can see similar trends of statistical decline (or lack of improvement that we'd expect) in scoring/efficiency/team offense in other players throughout history. We see plenty of scalability concerns with too many ball dominant players join together. A) The media almost universally considered this year's Lakers favorites to come out of the west (I think 15/17 ESPN writers from memory?), but Westbrook's poor portability was a poor fit with LeBron's okay portability. There are certainly injury/age concerns, but in the samples we saw, there were clear diminishing returns for B) the recent 'superteam' Nets, C) the Nash/Kobe/Dwight Howard Lakers, D) the Shaq/Kobe/Malone/Payton Lakers,E) and the Olajuwon/Barkley/Drexler/Pippen Rockets... all considered super teams and clear-title contenders that blew up. F) Chris Paul and James Harden showed some decline, though the Rockets were smart in their coaching and maximized their time apart. G) Similar studies have been done on teams with less portable sub-all stars and low all-stars (rather than all-nba/MVP level players) and they find similar declines in players perceived as not portable.

Team-building: You mentioned in your first post that smart team building should be able to avoid cases of diminishing returns. Then why do we keep seeing this throughout history? To me, when teams are trying to build title-contenders, they know they have to add value. They can't just have their superstar alone. So there's a spectrum of options. One extreme is Strategy A:, we have a single star with an army of perfectly-fitting role players (who can still be sub-all star level). Another extreme is Strategy B: load up on as much top-end talent as you can and hope the depth doesn't matter. Of course, there's a wide middle ground between the two. I'd argue that it's quite hard to do Strategy A.... not that it's never been done, but it's taken a rare caliber superstar with sufficiently well-fitting teammates (on offense or defense), with sufficiently vulnerable opponents. In most cases, you're going to need at least a second all-star level talent. And since I've shown you can get diminishing returns with even 2 stars who aren't that terrible with scalability, then I'd argue scalability should be a factor to consider, not just something to diminish as "a worse case scenario for on ball overlap".

RE magic vs curry results

I would point out magic didnt have anyone on the offensive level of durant (worthty is great but not quite as much) in 87, and kareem offense in 87 is good but not the kind of monster he was before that you could rely on. In sone ways he was offensively a bit like klay in that he played a smaller than the main star but still importan offensive role and could go off in a whim still

But even if we consider 87 and 2017 rosters comparable (they are close-ish regardless) curry has exactly 1 run at that level and didnt come even close to repeat it in 2018 or in the non durant years, and his 2016-2017 regular season offense was not ever close to being repeated in 2015,2018, 2019, 2021 or 2022

This gets into a philosophical question here, magic has more ultra elite offenses in both playoffs and reg season than curry so if we think a greater scope look and use multiple years to get a idea of what the "real" magic or curry offensive impact was, magic peak cones ahead

If we are more strict with the question and look at the peak years in a vacuum (which is also perfectly valid and can be done when there is somethingh unusual about that year that "explains" the outlier result) curry comes ahead

So my question is....what do you think was curry doing difgerently in 2017 than every other year even with durant where he didnt come close to those results?

Was the 2017 result really because curry was worth like 5 points more than in 2018? Or is it more likely that the whole team clicked and peaked together in a way that us hard to replicate (kinda like the 2001 lakers) and their real level is more like what they showed in 18 and 19?.

Cause here is the think, we have seen players like nash or lebron lead even better offense results than 17 warriors without as much ceiling raising praise

If in the same era lebron who some consider a weaker and less portable offensive player than magic can lead a 16-17 cavs offense which was slightly more dominant than the peak curry 17-18 warriors which had more talent .... are we sure magic couldnt replicate sonethingh close to it?

I wouldnt think lebron in 16-17 was multiple points more valuable offensively than 2018 lebron in a vacuum, but the talent and fit of the team was such that the result was much greater and o honestly see sonethingh sinilar with the 17 warriors vs other durant versions of the team and with 16 curry vs other durant less versions of regular season curry

Curry was not a less talented shooter or less smart player in 2021 than 2016 but he never replicated that regular season impact and outlier result

I'll try to be brief, since we've discussed some of this before. We may just not agree on some points, which is ok!

First question, why was 2017 better than 2018 or 2019? Like my comment with 70sFan, to me Curry has a medium-short peak (~2-3 years). His best 2 years were 2016 (when healthy) and 2017 (after adjusted to KD), with arguments for adding 2015 or 2018. He got close in 18/19/21/22, but wasn't quite the same as his true peak.

And to be clear: a 2 year peak is not too short to consider for this project. Shaq had a 2 year peak, and he was voted 3rd. Nobody cared about Shaq's worse performance in 1999 or 2002 when rating 2000/2001. Duncan had a 2 year peak and he was voted in already. Heck, if you're saying peak years need to be 3+ years in-a-row, people might need to be lower on Lebron's peak (12-13 is good, but 11/14 << 09), Wilt's peak (67/64 >> 65/66/68), Kareem (77/74 >> 75/76/78), Bird (86 >> 85/87), and 04 KG (04 >> 03/05). If we add context for Curry:
-Unlike any other player at this level, Curry got injured during the playoffs of one of his 2-year peaks (Walton and Kawhi share this pattern but their peak is lower).
-Unlike any other player in the top... 20 peaks? 30 peaks?... Curry had an all-time player join his team during his 2ish-year peak.

So why did Curry get worse after 17? Well, he was transitioning from his peak to his prime, just like everyone else does. The only reason to be lower on Curry's 2-year peak is injury risk (if you value risk over what happened, since he was healthy in 2017) or if you really care about regular season value without valuing context or goodness (which should make you lower on 16 LeBron, 94/95 Hakeem, and 03 Duncan, for example).

Second question, could LeBron lead as-good an offense as Curry? With perfect fitting role players and all-stars, it's possible Magic/Nash/LeBron would be able to lead a better offense than +8.1 rORTG in the regular season (16 Warriors, 3rd all time) or +11.4 rORTG in the playoffs (17 Warriors, 8th all time, min. 10+ games). But I absolutely don't think any of them could lead a better offense on those teams than Curry did. And I have serious doubts that either Magic or Nash could have the scalability to maintain that high an offense while playing with good enough defensive players to maintain a -5.8 rDRTG in the playoffs (17 Warriors, top 2 in league).

__________
Quick tangent to see if I can understand your point of view better (no worries if you're not interested): you've cited the Warriors worse performance without Klay in ~19 finals, 20, and 21, and the Warriors improved performance near Klay's return in 22 as reasons for being lower not just on Curry's prime, but his peak as well.

Do you see any similar arguments as compelling reasons for being lower on other players? For example:
-Kareem couldn't win without a Top 3-All Time point guard. He never won without Oscar or Magic.
-Jordan never won without a Top 5 Coach all time and one of the GOAT sidekicks in Pippin. The majority of people think he was much better in 88 or 89 than any of his second three-peat years. Yet he couldn't win in those 88-89 years without Phil Jackson/Prime Pippin.
-Shaq never won without a Top ~5 shooting guard of all time. He never won without prime Kobe (+Phil Jackson) or prime Wade.

I don't, but just wondering if you see any inconsistency with the Curry/Klay argument and these examples.

[DraymondGold]

I forgot but you mentioned you use the gitlab data for RAPM right? I’ve heard the Gitlab data in general is really weird, I think it was either eminence or ceiling raiser that mentioned it

This doesn’t change the argument for Curry because his whole season looks similarly impressive in other samples, but just worth mentioning

Almost all of my stats have been Goldstein RAPM, which is usually the gold standard for RAPM (along with Engelmann). I just used Gitlab once when someone asked for separate regular season / playoff values. But I appreciate the info about it! I didn't know it had that reputation

Year sample: Here's the problem with 3-year playoff samples for Curry. Any way you slice it, either 33% or 67% of his 3-year "peak playoffs" are limited by injury. Most players true "peaks" are somewhere around 1-4 years, with the amount of decline on either side varying. Going by CORP (a consistent estimate for goodness, though obviously not perfect), here's a few examples
Type 1) Long Peak and prime not far off: Bill Russell has a consistent 3-year peak with almost no drop off on either side.
Type 2) Long Peak but less good prime: Jordan has a consistent 3-year peak and a larger (but still not massive) drop off on either side.
Type 3) Medium peak less good prime: Duncan has a consistent 2-year peak with a large drop off on either side.
Type 4) Short peak less good prime: Shaq had an inconsistent 2 year peak (where 1 year was better than the other), with a large drop off on either side.
You might disagree with the example players or the stat, but you get what I'm going for. Under your argument, Magic falls into type 1: long peak and his prime is fairly close. Which is great!

But what do we do with shorter 2-3 year peaks that have a drop-off? This group didn't dock Shaq at all for having a short peak -- he was voted in third! Nobody used Shaq's 1999 playoffs or his 2002 playoffs to say that 2000 wasn't as good. This group also didn't dock Duncan for having a short 2-year peak and a clear drop off after. He was voted in over Hakeem!

Where does that leave us with Curry? What if he falls into Type 3 or Type 4? Well, if he had a 2-3 year peak with a drop off after, then ~2/3 of his best years were limited by a playoff injury! So if we take a 3-year playoff-only average to estimate 2017, then we're docking him in a year that he's healthy for the fact that he wasn't healthy in other years. If we do a 5-year playoff-only average for 2017, we're not only docking him for injuries when he was healthy in 2017, but we're also docking him for his worse prime years vs someone who had a better prime (but not necessarily a better peak).

Now I understand if people are concerned that 2017 Curry's a health risk, even if he stayed healthy. I uncertainty if there's concern for projecting what a healthy 2016 playoffs or a healthy 2018 playoffs might look like. But it's not like he never reached 2017-playoff impact levels ever... he reached them when healthy in 2016 and at times in 2015. And it's not like there isn't uncertainty for other players... every other player we're considering today has far fewer impact measures (and performs worse than Curry in the 1/2-year samples we do have). That's at least how I see it, but like I said, I can understand people's concerns.

In this case specifically, and also perhaps Duncan's case, theirs good reason to believe their 1 year peaks are actual legitimate outliers in their career. For Shaq, it was very evident that he stepped up his defensive effort while maintaining his insane offense and took conditioning/training more seriously in 2001 moreso than any other year. He genuinely put his best foot forward and actually gave his all which he readily admitted he didn't do for most of his career. And as such he put together by far the most complete year of his career. From both film and what has been written about 2001 Shaq, everything points to it being a legitimate outlier season for him.

I appreciate the consistency with Duncan! For Shaq, I imagine you're talking about 2000 (since that was his best defensive year), but I get what you mean. Perhaps some people are consistent with downgrading 2 year peaks vs 3/4 year peaks. I just see far more doubters with Curry's 2-year peak than I do with these two, despite the evidence that Curry's peak is in the same tier (with arguments over them).

It's possible in your definition of scalability/portability (e.g. how players value changes fitting on any other team), Magic fairs just as well as Curry. But as teammates get better, certain skills are going to slightly decrease in value.

For example, if we have lots of ball-dominant iso scorer on a team, there won't be as much time or opportunities for these ball dominant scorers vs if each iso scorer had their own offense they could lead alone. Someone's scoring value is going to have to decrease. But not all skills decrease as much as ball-dominant iso scoring when surrounded by better teammates. This skill analysis is how I tend to evaluate portability/scalability.

So what skills retain the most value with good teammates?
Scalability tier: Spacing ~ Finishing ~ Passing > (non spacing/passing) Creation > Iso Scoring
Scalability tier (other skills): Off ball > on ball. Defense is also very scalable (though off-ball defense > on-ball defense).
*Qualifier: this is not to say high portability skills are inherently more valuable on a good team... it's just that they lose less value when in a ceiling-raising role vs a floor-raising role. For example, it's possible that spacing is worth +0.25 on a bad team and+ 0.5 on a good team vs iso scoring which might be +1 on a bad team and +.75 on a good team.

That's interesting, do we have any evidences that passing is less scalable than spacing, or is it strictly philosophical approach? I don't question it with bad intentions, I find philosophical view very important in such discussions.

I have passing in the same tier as spacing/finishing in terms of scalability (all over iso scoring for example -- that's what I tried to show with the "~" sign vs the "<" sign).

But philosophically (without much metrics to back it up), I do have slight preferences for the scalability of spacing and finishing over passing, even though it's close. To me, it comes down to the scalability advantage of off-ball vs on ball moves, and passing tends to be more on-ball than the other two. Still, I am high on the scalability of what people call "off-ball passing" -- it's mostly just my scalability concerns with ball-dominant passing that puts it below finishing or especially spacing.


I don't have too much time to answer the rest of the points today (I wish! perhaps I'll edit in comments later), but to summarize:
-Re: Magic's rim finishing: I don't have any specific data for Magic's finishing, so sorry if it sounded like it! I was trying to suggest the general scoring gap + the data we have to show Curry's all-time point guard skill as a finisher + the film/eye-test of their rim-attempts makes me comfortable putting Curry a tier over Magic at rim scoring. But I don't have any data specifically on Magic's attempts Let me know if you think this is underrating Magic!
-Re: LeBron, that is a can of worms having Magic over LeBron for offense! That said... I don't exactly hate this take
-Re: Magic's BBIQ being a scalable skill... that's an interesting point, I hadn't thought of that There are some cases where all-time great BBIQ players still aren't perfectly scalable (LeBron comes to mind), but perhaps this should make me higher Magic's "ceiling" in terms of scalability

[OhayoKD]

Not too much to add on your comments about Peak length or scoring -- everything sounds reasonable there!

2. Creation: Magic > Curry, but Curry closes much of the Gap
Does Curry's superior screening single-handedly close the creation gap? Of course not But it's not about Curry's individual off-ball skills... it's about the culmination of his off-ball creation closing the creation gap far more than Magic can close the scoring gap.

Yeah, just to be clear - I didn't want to say that Curry's off-ball creation doesn't matter here, only that his screen setting isn't that important in this discussion

I think we have to ask ourselves: Why does Curry consistently create better shots for his teammates and improve his teammates' efficiency more than older LeBron, Harden, Jokic, Luka, Westbrook, and older Chris Paul, whether we're looking at a 1-year peak or a 5 Year prime? Where does this playmaking come from?

Curry's an all-star level passer, but it's clearly not passing. So that's where the culmination of all the little forms of creation come into play, where Curry is consistently all-time to GOAT level in these skills. He's constantly near the top of the league in guards' Screening Assists, at just under 1 per game. He's constantly near the top of the league in Secondary/Hockey Assists, at just over 1 per game. And per manal tracking, he has multiple off-ball movement assists and gravity-dragging assists per game. This says nothing about the subtler cases where Curry's presence makes it harder for opponents to double, help on, or close out on Curry's teammates.

Thanks for the numbers, such a shame that we don't have similar ones for Magic. I'd guess that Johnson is the kind of player that would thrive by these criteria, he turned limited roleplayers into gold.

I won't deny, in recent years I got a huge appreciation for Curry's off-ball game. It's really a game changer in a lot of games.

Per my film study earlier in this project: in the 2017 Finals, Curry drew the primary attention of at least 2 defenders on 62% of possessions where he was involved (34/55), and his teammates' points were made easier by this 89% of the time (34/38 points benefited from the attention Curry drew).

That's really cool data and these numbers are absurd I know that not all teams defended Curry in similar way (Cavs were extremely agressive at doubling Curry), but it's still ridiculous number. I'm pretty sure Magic wasn't doubled nearly as often, so it should be taken in consideration.

Agreed, I'd love to get these numbers for other players like Magic. And I'm glad you like my film data! I agree, the Cavs were extremely aggressive at doubling Curry, probably more so than other teams. But those are also the series that people quote when saying KD > Curry in those years, so it's worth considering context at least so we can better understand the public narratives. I can't help but think people quoting the finals stats are missing this subtler context. To me, the impact still remains when you double him -- it just turns into off-ball creation instead of scoring, which is far harder to pick up with the "untrained eye-test" or with traditional box stats. That's part of the reason why I think the plus-minus numbers for Curry are capturing something real.

It's also worth noting, Curry's traditional box-score numbers when you don't double him shoot up. Curry faced as much single-coverage in the 2022 Finals as he did against any other major team (at least to my eye/memory), and despite the help defense being all-decade level and despite going against the supposed DPOY in man coverage, his scoring shot up and suddenly the public/media are saying "hey this Curry guy's impact is pretty big"

To me, this the fundamental conundrum opposing defenses face against Curry. Put your best man defender on him and double him, and Klay/KD/Wiggins/Poole are getting easy poorly-contested/open shots against worse man defenders. Don't double him, and Curry goes on massive scoring tears that break the game.

Per NBA Tracking Data: In the 2018 Finals, Curry received double teams 2000% more (that's two-thousand times more) than KD.
Additional Film Study here:

Spoiler:

KD pre-Curry goes until 1:45. Skip to 1:45 to see Curry's impact.
[url]

Watch on YouTube

[/url]

And remember: None of these examples are captured by the traditional box score, so people who just look at assist numbers or box-score only metrics are likely underrating Curry's playmaking. But they would show up in more advanced stats. For example 2017 Curry improved his teammates' scoring efficiency by almost twice as much as the next best playmaking star in the league in LeBron.

Not trying to take away any credit from Curry, but Cavs strategy was ridiculous

Just to clarify (sorry if my post was ambiguous!), Curry improved his teammates' scoring efficiency by almost twice as much as the next best playmaking star throughout the entire 2017 regular season. So this applies even when Curry's not facing the Cavs' somewhat-ridiculous level of defensive attention. Just to make it more explicit, here were the numbers I meant (same stats I showed after homecourtloss' question):

1 Year Peak: Curry +7.3% (1st in league) >> older LeBron +3.9% (2nd in league) > peak Westbrook +2.5% (3rd in league)
[Metric: teammates' shooting percentage improvement when a star is on court vs when they're off. Sample: among top players in 2017]
5 Year Peak: Curry +0.07 (1st in league) > Trae Young +0.06 (2nd in league) > Nikola Jokic +0.06 (3rd in league) > LeBron James +0.05 (4th in league)
[Metric: teammates' increased pts/shot when a star is on vs off, aided by tracking data. Sample: 2018-2022]

I think if people aren't considering Curry a possible top 5 playmaking offensive engine of all time, they're seriously underrating Curry. To be clear, I still have Magic as the superior playmaker, largely from his volume advantage as a playmaker. But Curry is Tier 1 All Time in terms of playmaking efficiency, and I think these advanced stats/film analysis support that the playmaking gap is smaller than the scoring gap, at least to me.
Sources:

Spoiler:

1. 1-Year peak Teammate shot improvement: https://fivethirtyeight.com/features/the-case-for-stephen-curry-mvp/
2. 5 Year Prime teammate shot quality improvement: https://synergysports.com/explaining-synergy-shot-quality/

Read on Twitter

3. My 2017 Finals film study: https://forums.realgm.com/boards/viewtopic.php?p=100386706#p100386706
4. 2018 Finals Tracking Data: NBAlogix (paywall) / Clutchpoints
5. Curry vs LeBron efficiency stats are in source 1

I'm not sure about top 5, but he's definitely a very strong candidate. The only ones I'm confident taking over him are Magic, Nash and LeBron. I'd consider others like Oscar, but without more film studies I'd give Curry the edge.

Yep, he might not be consensus top 5 (Magic, Nash, and LeBron are the usual suspects at the top), but I definitely think he should be in at least in contention for top 5.

3. Overall Impact: Curry > Magic
Good point that we have to look at the whole package, not just individual skills! To me, that's what I'm doing when I'm looking at the metrics. For example, if we look at their ScoreVal + PlayVal (to approximate combined scoring and playmaking value):
3-year Regular Season: 15-17 Curry's 4.2 >87-89 Magic's 3.3.
Peak Playoffs: 17 Playoff Curry's +4.3 > 87 Playoff Magic's +3.7.
And like I've mentioned, the other 1-3 year metrics favor Curry (see below or my Ballot on page 1). I also still feel confident about this after the film study, though we do seem to disagree here. Would you recommend a different way to look at their wholistic value?

That's interesting idea, I decided to play with the numbers:

2015 Curry: 4.0 in RS, 3.0 in PS
2016 Curry: 5.2 in RS, 2.1 in PS
2017 Curry: 3.3 in RS, 4.3 in PS
2018 Curry: 3.8 in RS, 3.3 in PS
2019 Curry: 2.5 in RS, 2.4 in PS

Average: 3.8 in RS, 3.0 in PS

1987 Magic: 3.4 in RS, 3.7 in PS
1988 Magic: 2.6 in RS, 3.4 in PS
1989 Magic: 3.9 in RS, 3.2 in PS
1990 Magic: 3.8 in RS, 4.0 in PS
1991 Magic: 3.5 in RS, 3.3 in PS

Average: 3.5 in RS, 3.5 in PS

As you can see, Curry has the edge in RS, but Magic's advantage is quite notable in the PS. You compared 3-year RS samples, but only 1-year playoffs, which gave the impression that Curry has the advantage in both categories. The truth is that this is by far the best Curry postseason run (without counting 2022, as I don't have numbers for it) and every other Curry's run is lower than Magic's top 3 runs. It's also important to note that Curry's 2017 RS is notably worse than Magic's best ones.

Maybe it has something to do with injuries, but Curry didn't have any notable ones in 2015 or 2019 and he still didn't look on prime Magic level.

I think that some people think that without the injury, Curry would have had 2017 level run in 2016, so they give him the benefit of doubts, but I'm not sure we can say that for sure.

Glad you brought this up! And you're right, I only included 1-year sample for playoffs.

Injuries: Here's the problem of course, which you mention: in both surrounding years to his playoff peak, Curry was injured enough to miss 6 games! So his playoff numbers will seem lower if he isn't healthy by the time he's playing (2016), if he doesn't have time to get back up to playoff speed / playoff rhythm (both 2016 and 2018), and if he doesn't get to boost his numbers in the weaker first/second round (both 2016 and 2018).

Injury affects on numbers: KD's on/off data in the 2018 playoffs look a lot closer to Curry's, but there's a problem -- KD gets a huge boost from the first round matchup, which Curry didn't get. The reverse is in 2019, where KD misses the harder matchup while Curry gets the harder matchup. Looking at just games where they both played, we get back to the normal pattern of Curry clearly outperforming KD. I wonder whether something similar might be happening here, where most people get to boost their scoring/playmaking in the first round, while 2018 Curry only faced harder teams. I wonder if something

Defensive attention: it's also worth noting that PlayVal uses box-score stats, so it wouldn't capture the off-ball creation we discussed previously. Who knows how much playmaking value isn't captured by PlayVal due to it being off-ball (for either Curry or Magic), but that may be a factor.

Year sample: Here's the problem with 3-year playoff samples for Curry. Any way you slice it, either 33% or 67% of his 3-year "peak playoffs" are limited by injury. Most players true "peaks" are somewhere around 1-4 years, with the amount of decline on either side varying. Going by CORP (a consistent estimate for goodness, though obviously not perfect), here's a few examples
Type 1) Long Peak and prime not far off: Bill Russell has a consistent 3-year peak with almost no drop off on either side.
Type 2) Long Peak but less good prime: Jordan has a consistent 3-year peak and a larger (but still not massive) drop off on either side.
Type 3) Medium peak less good prime: Duncan has a consistent 2-year peak with a large drop off on either side.
Type 4) Short peak less good prime: Shaq had an inconsistent 2 year peak (where 1 year was better than the other), with a large drop off on either side.
You might disagree with the example players or the stat, but you get what I'm going for. Under your argument, Magic falls into type 1: long peak and his prime is fairly close. Which is great!

But what do we do with shorter 2-3 year peaks that have a drop-off? This group didn't dock Shaq at all for having a short peak -- he was voted in third! Nobody used Shaq's 1999 playoffs or his 2002 playoffs to say that 2000 wasn't as good. This group also didn't dock Duncan for having a short 2-year peak and a clear drop off after. He was voted in over Hakeem!

Where does that leave us with Curry? What if he falls into Type 3 or Type 4? Well, if he had a 2-3 year peak with a drop off after, then ~2/3 of his best years were limited by a playoff injury! So if we take a 3-year playoff-only average to estimate 2017, then we're docking him in a year that he's healthy for the fact that he wasn't healthy in other years. If we do a 5-year playoff-only average for 2017, we're not only docking him for injuries when he was healthy in 2017, but we're also docking him for his worse prime years vs someone who had a better prime (but not necessarily a better peak).

Now I understand if people are concerned that 2017 Curry's a health risk, even if he stayed healthy. I uncertainty if there's concern for projecting what a healthy 2016 playoffs or a healthy 2018 playoffs might look like. But it's not like he never reached 2017-playoff impact levels ever... he reached them when healthy in 2016 and at times in 2015. And it's not like there isn't uncertainty for other players... every other player we're considering today has far fewer impact measures (and performs worse than Curry in the 1/2-year samples we do have). That's at least how I see it, but like I said, I can understand people's concerns.

B) When you say "Magic... [is] just as spectacular" by the data... are we sure that's true?

Pure Impact Metrics: Curry >= Magic, though we're missing some of Magic's data. Curry's playoff-only PIPM is higher, and this applies for a 3-year playoff sample (17-19 Curry's 8th all time > 15-17 Curry (with injury) 17th all time > 87-89 Magic's 18th). Curry's estimated prime WOWY is higher (1st all time > 5th all time). 85 Magic's 41-game sample regular season RAPM edges out 2017 Curry's regular season, though 88 Magic's 54-game sample regular season RAPM falls behind 2017 Curry's regular season, and many people argue Curry's 2016 regular season was better.

I think WOWY is way too noisy to compare exact values. Curry and Magic both looks like the GOAT-level impact players from WOWY studies and I don't think we can conclude much more from that.

I don't think we can compare RAPM across seasons, isn't that incorrect way to use it?

Re: WOWY, it's definitely noisy, but it is stabilized in that I used full 10-year samples for Curry and Magic.

For RAPM, good question! We can look at how dominant people were relative to their league, which are the numbers I gave (and have been using for these playoffs). With these in-era numbers, both Magic (in limited sample) and Curry outperform basically everyone else people are considering. Or, we can normalize across history, to get some sort of cross-era time machine numbers, which we can do until 1998, but we can't do for pre-97 Historic RAPM data (at least as I understand Squared2020's excellent RAPM research -- happy to be corrected here if I'm wrong!).

Box-metrics: Curry >> Magic. What about box models of plus minus data? Curry again has the advantage, and the stronger advantage in the playoffs. 2017 Curry's higher than 1987 Magic in postseason Backpicks BPM, BR’s Postseason BPM, and WS/48. (I can't seem to convince people that PER is a bad stat... so if you can't beat em, join em: 2017 Curry's PER is higher than 1987 Magic's PER too). Magic never beat Curry's 2017 playoff numbers in Backpicks BPM, BR’s Postseason BPM, and WS/48, or PER.

And remember, Curry is Top 2 All Time in regular season AuPM, postseason AuPM, on/off, ESPN’s RPM, and RAPOR +/-. I think your point that Magic's non-peak Prime years may surpass Curry statistically certainly has a case. But peak for peak, in 1/2/3 year samples, I'm not sure the data supports the idea that Magic is just as spectacular as Curry.

Yeah, I guess I still have the problem with how massive outlier Curry's 2017 postseason is compared to the other surrounding years (even excluding 2016).

Anyway, that's why I have Curry > Magic on offense. And since the film study shows they're similar on defense, I think this explains why Curry has the overall higher impact metrics. Let me know where y'all disagree!

I want to thank you for this very interesting conversation, they are far more important than the results of voting. I am still on Magic's side here, but you gave me another look at Curry's seasons in retrospect and I found more appreciation for what he did.

I've enjoyed it too! Your work on Kareem definitely gave me more appreciation for him, so thanks for the great discussions!

is "10 year wowy" really all that useful? Like you'd basically just be looking at a couple games per season right?

[Dr Positivity]

1. Curry 2015 (b. 2017 c. 2019) - the best PG skillset of all time for me

2. Giannis 2022 (b. 2021 c. 2020) - I might give slight edge to other players in terms of how good they were for their season, but I think Giannis makes up for it by playing in the most modern time period

3. Jokic 2022 (b. 2021 c. 2020) - Essentially the same argument, if the other players are better for their seasons it's not by that much, I think I favor the ultra dominant modern one. He is worse on defense than other options but arguably more dominant on offense for his position than say Dirk. Willing to be talked out of it since he's not likely to get in for at least 4-5 more threads

[falcolombardi]

...

Excellent post.



__________

RE: Big men defense

Do you consider perimeter defense for centers a bigger weakness than a perimeter player who gets picked on for being a bad individual defender, is too small or messes the rotations off ball? Cause a lot of offensive perimeter guys have one or many of those limiations (lillard, nash and westbrook come to mind respectively) and teams still played them

A center who is bad on the perimeter but solid in the paint (the more important part of a center job) can be "hid" by playing a scheme that avoids him switching on guards and teams that go with multiple big (like, lets say boston, milwaukee) manage incresible defense by going big even when guys like lopez and horford are not "perimeter centers" un defense by any definition

Great question! I still consider big man on average to be more valuable on defense than perimeter players of course. The point I was specifically arguing for was value today relative to a ~15 years ago relative to ~25 years ago. Big men are far bigger perimeter defensive liabilities today than they were back in the day; meanwhile I wouldn't say perimeter players have lost much value to being liabilities in the post.

If you forced me to pick who's the bigger liability on average today, I am open to the idea that big men are bigger perimeter liabilities in today's era than perimeter players are liabilities in the paint. Why? It personally feels harder to pre-switch the pick and roll or continuous perimeter screens than it is to pre-switch to prevent a big men on smaller defender. Some of this comes down to speed of action -- the faster the action, the harder it is to pre-switch, and it's a lot easier to have quick screening action on the perimeter vs in the paint. The cost of being slower on a screen on the perimeter is also at an all-time high. Meanwhile, help defense is far easier in the paint vs the perimeter. Some examples today might be various star drop bigs in playoff series, like this year's Nuggets vs Warriors, or 21 Jazz vs Clippers, or 19 Jazz vs Houston.

Regardless of which liability is bigger, I still feel big men on average add more value with rim protection than perimeter players do with perimeter defense.

RE:ON-BALL overlap

While it is theorically true, you are taking a case like 2012-2014 heat, a team where halfway through that run wade started to decline with injuries, as evidence

If you look at their offense 2012-2013 before the pkayoffs vs after that, the splits are night and day

You are taking a worse case scenario for on ball overlap here because lebron already can create elite offense with shooters/finishers and without another star so the margin to improve is smaller and then combining him with an athletic slasher/cutter without the athletism (wade post injuries) who is a weaker spot up shooter so he can be ignored off ball and is not a big positive on ball either

If wade played with curry instead i am sure you would see a similar split post injury for wade/curry, is the equivalent of durant playing with a broken hand that doesnt let him jumpshot with the defense knowing it and blaming curry/durant fit for the bad results

Is it really worst case scenario? Sure, Wade declined, but LeBron was at his peak, and I'd have plenty of players as less portable than LeBron. Even if Wade was declining, he wasn't only worth "2 wins" in his older age... that can only be explained with diminishing returns. And even if he's worse, he's certainly still a better player in raw value than whoever they had coming off the bench for him. So why does LeBron's scoring and shooting numbers and the team's offense? The only explanation I can come up with is diminishing returns with better teammates (which is another name for worse fit with better players... which is another name for scalabilty/portability).

More Poor Scalability Examples: And like I mentioned, you can see similar trends of statistical decline (or lack of improvement that we'd expect) in scoring/efficiency/team offense in other players throughout history. We see plenty of scalability concerns with too many ball dominant players join together. A) The media almost universally considered this year's Lakers favorites to come out of the west (I think 15/17 ESPN writers from memory?), but Westbrook's poor portability was a poor fit with LeBron's okay portability. There are certainly injury/age concerns, but in the samples we saw, there were clear diminishing returns for B) the recent 'superteam' Nets, C) the Nash/Kobe/Dwight Howard Lakers, D) the Shaq/Kobe/Malone/Payton Lakers,E) and the Olajuwon/Barkley/Drexler/Pippen Rockets... all considered super teams and clear-title contenders that blew up. F) Chris Paul and James Harden showed some decline, though the Rockets were smart in their coaching and maximized their time apart. G) Similar studies have been done on teams with less portable sub-all stars and low all-stars (rather than all-nba/MVP level players) and they find similar declines in players perceived as not portable.

Team-building: You mentioned in your first post that smart team building should be able to avoid cases of diminishing returns. Then why do we keep seeing this throughout history? To me, when teams are trying to build title-contenders, they know they have to add value. They can't just have their superstar alone. So there's a spectrum of options. One extreme is Strategy A:, we have a single star with an army of perfectly-fitting role players (who can still be sub-all star level). Another extreme is Strategy B: load up on as much top-end talent as you can and hope the depth doesn't matter. Of course, there's a wide middle ground between the two. I'd argue that it's quite hard to do Strategy A.... not that it's never been done, but it's taken a rare caliber superstar with sufficiently well-fitting teammates (on offense or defense), with sufficiently vulnerable opponents. In most cases, you're going to need at least a second all-star level talent. And since I've shown you can get diminishing returns with even 2 stars who aren't that terrible with scalability, then I'd argue scalability should be a factor to consider, not just something to diminish as "a worse case scenario for on ball overlap".

RE magic vs curry results

I would point out magic didnt have anyone on the offensive level of durant (worthty is great but not quite as much) in 87, and kareem offense in 87 is good but not the kind of monster he was before that you could rely on. In sone ways he was offensively a bit like klay in that he played a smaller than the main star but still importan offensive role and could go off in a whim still

But even if we consider 87 and 2017 rosters comparable (they are close-ish regardless) curry has exactly 1 run at that level and didnt come even close to repeat it in 2018 or in the non durant years, and his 2016-2017 regular season offense was not ever close to being repeated in 2015,2018, 2019, 2021 or 2022

This gets into a philosophical question here, magic has more ultra elite offenses in both playoffs and reg season than curry so if we think a greater scope look and use multiple years to get a idea of what the "real" magic or curry offensive impact was, magic peak cones ahead

If we are more strict with the question and look at the peak years in a vacuum (which is also perfectly valid and can be done when there is somethingh unusual about that year that "explains" the outlier result) curry comes ahead

So my question is....what do you think was curry doing difgerently in 2017 than every other year even with durant where he didnt come close to those results?

Was the 2017 result really because curry was worth like 5 points more than in 2018? Or is it more likely that the whole team clicked and peaked together in a way that us hard to replicate (kinda like the 2001 lakers) and their real level is more like what they showed in 18 and 19?.

Cause here is the think, we have seen players like nash or lebron lead even better offense results than 17 warriors without as much ceiling raising praise

If in the same era lebron who some consider a weaker and less portable offensive player than magic can lead a 16-17 cavs offense which was slightly more dominant than the peak curry 17-18 warriors which had more talent .... are we sure magic couldnt replicate sonethingh close to it?

I wouldnt think lebron in 16-17 was multiple points more valuable offensively than 2018 lebron in a vacuum, but the talent and fit of the team was such that the result was much greater and o honestly see sonethingh sinilar with the 17 warriors vs other durant versions of the team and with 16 curry vs other durant less versions of regular season curry

Curry was not a less talented shooter or less smart player in 2021 than 2016 but he never replicated that regular season impact and outlier result

I'll try to be brief, since we've discussed some of this before. We may just not agree on some points, which is ok!

First question, why was 2017 better than 2018 or 2019? Like my comment with 70sFan, to me Curry has a medium-short peak (~2-3 years). His best 2 years were 2016 (when healthy) and 2017 (after adjusted to KD), with arguments for adding 2015 or 2018. He got close in 18/19/21/22, but wasn't quite the same as his true peak.

And to be clear: a 2 year peak is not too short to consider for this project. Shaq had a 2 year peak, and he was voted 3rd. Nobody cared about Shaq's worse performance in 1999 or 2002 when rating 2000/2001. Duncan had a 2 year peak and he was voted in already. Heck, if you're saying peak years need to be 3+ years in-a-row, people might need to be lower on Lebron's peak (12-13 is good, but 11/14 << 09), Wilt's peak (67/64 >> 65/66/68), Kareem (77/74 >> 75/76/78), Bird (86 >> 85/87), and 04 KG (04 >> 03/05). If we add context for Curry:
-Unlike any other player at this level, Curry got injured during the playoffs of one of his 2-year peaks (Walton and Kawhi share this pattern but their peak is lower).
-Unlike any other player in the top... 20 peaks? 30 peaks?... Curry had an all-time player join his team during his 2ish-year peak.

So why did Curry get worse after 17? Well, he was transitioning from his peak to his prime, just like everyone else does. The only reason to be lower on Curry's 2-year peak is injury risk (if you value risk over what happened, since he was healthy in 2017) or if you really care about regular season value without valuing context or goodness (which should make you lower on 16 LeBron, 94/95 Hakeem, and 03 Duncan, for example).

Second question, could LeBron lead as-good an offense as Curry? With perfect fitting role players and all-stars, it's possible Magic/Nash/LeBron would be able to lead a better offense than +8.1 rORTG in the regular season (16 Warriors, 3rd all time) or +11.4 rORTG in the playoffs (17 Warriors, 8th all time, min. 10+ games). But I absolutely don't think any of them could lead a better offense on those teams than Curry did. And I have serious doubts that either Magic or Nash could have the scalability to maintain that high an offense while playing with good enough defensive players to maintain a -5.8 rDRTG in the playoffs (17 Warriors, top 2 in league).

__________
Quick tangent to see if I can understand your point of view better (no worries if you're not interested): you've cited the Warriors worse performance without Klay in ~19 finals, 20, and 21, and the Warriors improved performance near Klay's return in 22 as reasons for being lower not just on Curry's prime, but his peak as well.

Do you see any similar arguments as compelling reasons for being lower on other players? For example:
-Kareem couldn't win without a Top 3-All Time point guard. He never won without Oscar or Magic.
-Jordan never won without a Top 5 Coach all time and one of the GOAT sidekicks in Pippin. The majority of people think he was much better in 88 or 89 than any of his second three-peat years. Yet he couldn't win in those 88-89 years without Phil Jackson/Prime Pippin.
-Shaq never won without a Top ~5 shooting guard of all time. He never won without prime Kobe (+Phil Jackson) or prime Wade.

I don't, but just wondering if you see any inconsistency with the Curry/Klay argument and these examples.

First and foremost i wanna tell that this has been a very interesting post and the politeness with which you discuss thinghs is very cool

I hope i make my view here more clear

Now i get your point about 16-17 curry as his peak and everythingh else as his prime, if we combine the inpressive regular season of 16' with the "ceiling raising" of a more talented team in 17' then that is indeed a hell of a peak that i would feel comfortable-ish ranking above magic even from just a results perspective

I am not completely sold on it because i dont think peak magic got to play with anyone comparable to durant in offense (kareem was still good at 87 but not that good) so i mentally curve the best magic result vs the best curry result as similar given the talent gap (87 vs 17)

but when i look at surrounding seasons i am not comfortable assuming 16' curry would have had a comparable playoffs offense to the regular season one if not for the injuy, is possible but not to take for granted, specially because the issues we saw curry having vs cavs in 16 felt similar to those in 15 (+1 rel ortg vs cavs)

the cavs were a better defensive roster than their reg season once love and kyrie went down so i dont want to overstate the relevance of that +1 but both in 15 and 16 we see teams being able to slow down curry with agressive off ball defense and in 18 even with durant we see the warriors offense being stopped again below peak magic levels by a strong

That is what gives me doubt, the contrast between magic season is much lower, the variability less pronounced and the post season drop offs less common

Now if we approach their peak 1 year in a vacuum which is also valid and philpsophically consistent with a peaks project too so i have no real disagreement with your approach here for 16-17 curry

I know i may be coming across as a designated curry hater but i actually have him in my next probable 4 picks alongside garnett, bird and magic (and a tad above the walton, robinson, oscar, west, wade group) the fact i am not completely confident in magic>curry (although i lean towards it) or for that matter bird vs magic and bird vs curry is why i probably will skip this round since i cannot choose here

RE: wade and lebron fit

On this topic i think you may be underating the drop off from wade from the 13 season to the 13 playoffs, take a look at the boxscore stats drop even beyond volume (free throw rate, efficiency)

Wade (who i am very high on as you can notice from where i am considering him so this is not me taking the piss on him to elevate lebron) was, essentially a neutral at best player in those playoffs

The splits with wade on vs off are just mindblowing, the finals in particular. is not that heat didnt improve much adding wade, is that they became way worse when wade played that just doesnt happen cause portability, it happens when a player is actively a negative

He was a slasher who couldnt slash effectively, but was also a weak enough shooter to leave open. And with the physical struggle he was not even the usual lob/cutting/fast break threat. That is a recipe for disaster. Remember he was actively playing through pain that postseason too

The comparision with westbrook is even less fitting cause westbrook is actively a player in decline since at least 2019 but really 2018 when he left his peak (take it from the okc fan, it was hurtful as a westbrook fan to see through the last 5 years of gradual fall lol) his physycal and weirdly skill decline (he must be the only player to become a worse 3-point, midrange and free throw shooter with age) has been weird as hell

Yet the odd fact is that the warriors offense in 2021 was almost as bad as the lakers offense in 2022 despite much better health

and when this is pointed out as a example that even curry for all his portability needs a team constructed reasonably well around his strenghts to have strong results (the 2021 result) it is usuallt handwaved away cause warriors were playing oubre and other guys who dont fit curry and the warriors system

Exactlt the same thingh on-ball players get criticized for (having weaker results when they play in rosters that actively hamper their approach such as when you play a slasher alongside non shooters) is a excuse for off-ball ones

That the warriors awful offense in 2021 doesnt count cause they were badly constructed to fit curry and the team style in spite of curry being suposeddly all about fitting any roster and kind of talent

but that lebron struggles when he plays with a hurt player that hampers him (injured wade ) is cause he is not portable enough. I think the problem is that we still think of that injured (13 playoffs) or diminished (14 and forward) wade as THE wade. Not as a much diminished fascimile ala bulls jordan vs wizards jordan

Does what i am saying make sense? We can just go look into wade and lebron results in 2013 regular season and 2012 and even 2011 seasons and playoffs. There is a 2 season and half sample of lebron and wade being tremendly good together amd better thab lebron or wade alone but wade gets hurt and suddendly any lineup with him is struggling, with or without lebron on them

I am unsure why would we blame the portability rather than the injury considering all that?

were lebron and wade more portable the same season when wade was not hurt and they were going on 27 win streaks and having a 66 win season? Then suddendly had negative portability x 10 in the playoffs which coincidentally happened when wade was hurt and injured and playing througg pain and injury?

TLDR i wrote a lot and hope i wrote somethingh readable and coherent to express my thoughts, i thank you in advance for this nice discussion too and for your patience responding always in depth like you do

You are always a nice poster to discuss thinghs with even if the points of view are so opposite at times

[f4p]

Yet the odd fact is that the warriors offense in 2021 was almost as bad as the lakers offense in 2022 despite much better health

and when this is pointed out as a example that even curry for all his portability needs a team constructed reasonably well around his strenghts to have strong results (the 2021 result) it is usuallt handwaved away cause warriors were playing oubre and other guys who dont fit curry and the warriors system

Exactlt the same thingh on-ball players get criticized for (having weaker results when they play in rosters that actively hamper their approach such as when you play a slasher alongside non shooters) is a excuse for off-ball ones

That the warriors awful offense in 2021 doesnt count cause they were badly constructed to fit curry and the team style in spite of curry being suposeddly all about fitting any roster and kind of talent

exactly. steph curry has largely played on the most perfectly constructed roster ever over his prime and, when he looks good on that perfectly constructed roster, it's somehow taken as evidence that he will succeed on any roster. you couldn't construct a player in a lab to fit curry better than draymond green and if you were going to get a secondary offensive player, you would probably want a great off-ball player like klay thompson who fits into curry's preferred offense, maybe even one whose weak ball-handling ensures his ego won't get out of control and compete with the alpha for touches. and you'd want a coach who puts it all together.

when the warriors don't have a great offense under mark jackson, it's because mark jackson is dumb. when the warriors struggle in the 2015 finals, it's not because off-ball play and motion offense have their own drawbacks, just like any system. when they lose in 2016, it's injuries and not more of the same from 2015. when the 2018 team almost loses, it's because kevin durant decided to run an isolation play, a corrupted abomination of God's creation. how dare the warriors be expected to be resilient in the face of one of the great iso players of all time deciding to iso. when they barely beat the rockets again in 2019, with steph playing horribly, it's again because the king of scalability was being held back by durant. his success against portland without durant somehow doesn't show he couldn't quite scale with durant, despite having far less overlap than guys like lebron and wade, who might be the most overlapping star pairing ever. but also when the warriors go 1-4 against the raptors without KD, that also doesn't show that they needed KD.

and as you pointed out, when the 2021 warriors finish 22nd in offense, that can't be held against steph or indicate that maybe their offensive system catering only to curry might hold other players back. watching warriors fan react to oubre was fascinating. it was like it didn't occur to them that most of the nba is not draymond/iggy, guys who don't want to shoot and have their dribbling/passing/IQ sliders set to 99 to run complicated offensive sets. plenty of guys need to be set up and not have to think. even this year's warriors team only finished 17th in offense, even with klay back for about half the season and poole taking a huge step up. obviously some of that was poor health with the big 3 barely playing together, and also steph himself holding it back by struggling with his shooting, but it's still interesting that the king of gravity and scalability doesn't have a top 10 offense other than the 5 years you would obviously think of from 2015-19.

[MyUniBroDavis]

On the convo about 2021 being indicitive, their main issue was offensive rebounding and the games Curry missed, their effeciency was solid but I do think you do need multiple decent passers in a motion offense and they didn’t really have that outside of Curry and dray

Anyways, was gonna type this up for the next one but changed it around a bit

2017 Curry

Was thinking of putting Giannis above but I feel it was the perfect offensive performance on a great team, level of play it’s hard to beat but I don’t know if I’d pick Curry over giannis in a vacuum



2021 Giannis

I’m a bit conflicted here, but contextually I do think he should be on here already.

One thing I will say though is that I was looking through the retro player of the years for 2019 and 2020, while I thought he deserved a spot here, my assumption for why Giannis hasn’t been chosen yet or had much consideration was because people were lower on his offense than I was, I now realize it looks like people really didn’t rate him highly on defense?

The peak Duncan vs peak Giannis on offense thread suprised me a bit, I didn’t think it would be so clear cut, my assumption was that people weren’t high on his offense as much as his defense.

The basis of my argument comes from 2021 and 2022 giannis not being a different defender than 2019 or 2020 giannis, rather less effort in the regular season (and I know this past season they mixed up their screen coverages like Miami for postseason versatility), supported by his impact data in the playoffs, the bucks going from league average to probably best or near best in the league defensively in the playoffs

I don’t think 2021 or 2022 Giannis is intrinsically a different defender than his 2019 or 2020 versions, I do think that in general they took those defensive seasons off and ramped it up in the playoffs.

So my arguments on 2021 Giannis defensively will revolve around the impact of his defense in 2019, and in 2020 first and foremost

I wasn’t going to make much of a take about giannis’s defence mostly because I didn’t realize what the general opinion was vs I guess my opinion on his defence.

However, for this to make sense a base assumption is that 2021 playoff giannis on D is the same as 2019 or 2020 RS giannis on D, although there’s an argument for him being better than his 2019 self at least.

So ive kinda gotten the vibe that people (based off retro player of the year voting) didn’t think of him as a super clear DPOY those two years? 2020 I can understand because of ADs postseason exploits, but purely RS wise, those two were imo the best DPOY seasons post 2010.

I realize that might seem like a wild take (I honestly thought it was a pretty normal take) but It’s just that… it just seemed kind of obvious?

First of all, data when he did play vs when he didnt

NOTE: for games with giannis, I got it for 2019 but since it was the same as the on court percentage +_

2019
72 games
105.3 def rtg with giannis

10 games, 108.5 def rtg
(107.7 def rtg 8 games, 2 without starters)


- Avg off rtg faced (10 games) - 108.9
- Avg off rtg faced (8 games) - 108.5

2020
102.5 def rtg with giannis 63 games

108.6 (7 games)
108.5 (10 games, 2 in bubble 1 without starters)

- avg off rtg faced 109.2
- Avg off rtg faced 108.45

So in the 20 or so games that they played without giannis, the bucks were a slightly better than league average defense given their competition, taking only the 15 games their starters played or that weren’t in the bubble, they’re still slightly above average

Not the biggest of samples and there are a few outlier results both ways as is the nature of games in small samples, but still not insignificant. The bucks with giannis otoh, are a -5.2 defense in 2019 and a -7.7 defense in 2020.

Keep in mind budenholzer is weird with rotations, giannis only plays 30 minutes a game not because he’s unable to play longer but because budenholzer is known to be SUPER conservative in this regard

His on court and raw net rtg compared to other ATG defensive teams

https://syndication.bleacherreport.com/amp/2185159-ranking-the-nbas-20-best-defenses-of-all-time.amp.html

On court def rtg

2019 bucks
G 104.5 in 2019 (-3.7 net) (110.4 league avg)

2020 bucks
G 99.6 in 2020 (-7.9 net) (110.6 league avg)

1999 Spurs
David Robinson 92.0 (-2.4) (102.2 league avg)

2004 pistons
Wallace 94.8 (-2.0 net) (102.9 league avg)

2004 Spurs
Duncan 92.1 (-5.4 net) (102.9 league abg)

2005 Spurs
Duncan 94.5 (-8.9 net) (106.1 league avg)

2008 Celtics
Garnett 97.3 (-4.1 net) (107.5 league avg)

2014 pacers
Hibbert 98.9 (-2.9 net) (106.7 league avg)

In this comparison Giannis looks pretty solid, his 2019 season maybe a bit behind, his 2020 season fully deserving to be here.

I’m of the belief that top end defensive impact is limited in the past few years, the advent of three point shooting just means there are less shots to contest at the rim, and the pick and roll means especially in the playoffs you need versatility. Furthermore, even in the RS this leads to the defensive impact one individual can have to be a bit limited in comparison.

Kind of similar to how offensive numbers are inflated nowadays, defensive numbers are probably deflated (in terms of blocked shots and everything) because less blockable shots at the rim occur compared to back then (like it’s easier to block someone driving in a Congested paint than in a 5 out while ur drawn out to the three point line)

I think offensive impact isn’t as inflated maybe because general offensive “scheme evolution” and to a large extent more lax rules as well just means teams can do a bit better when their stars leave the court, but this pretty much has the opposite effect defensively

I think here’s where a bit of an issue comes in for me. Like how I believe it’s probably silly to say, oh devin booker averaged XXX he’s as good as Kobe, it’s also similarly strange for me to think that all the best defenders of today are clearly inferior to the defenders back then.

Or another way to say it, comparing the best defenders now to the best defenders back then, maybe I would get it more had TD and KG been so far head over shoulders above everyone else in defensive impact stuff, the first few Boston KG and some years of TD were, but as a whole while they stand out it’s not in the sense that they consistently dwarfed everyone else like currys offensive impact stuff I think, rather it seems there were usually a few guys that had really high impact in general

Another way to say it, I think the potential defensive impact today has lessened, and that especially on good teams if you transport a guy like Mutombo to todays game, he would not be AS impactful, despite his in eta impact based off rapm and what not making him seem like a (non russell) GOAT defender candidate with how much he dwarfed everyone else

Like, when we talk about how KG is perfect defensively today, I agree, but I don’t think that means he’s more impactful today than he was in the 2000s, I think this means he has less of a dropoff than some others would. This isn’t to say the others wouldn’t be absolutely fantastic but I think defenders have less impact in general today. I feel if you take a Rudy Gobert, whose absurdly dominant as an interior defender he’s probably gonna be a perennial DPOY candidate

Like he’s probably about as good as a mutombo, but even if we say a bit worse, who by some RAPM data consistently dwarfed the competition, and some of his post prime years were competitive or even better than some prime TD years, so I think peak Gobert is probably in the running for DPOY and in that category, even if he probably didn’t peak as high on that end (thats a whole can of worms)

The likes of TD and KG do dominate their era in metrics like defensive RAPM far more than Gobert does, although bigs don’t tend to dominate that as much as they did back then.

I used bball ref for the stuff above, admittedly just because it is far easier and older data seems hard to find.

Luck adjusted RAPM vs regular RAPM is a can of worms imo, luck adjustments generally work better in testing but for multiple years apparently they’re worse so take that with a grain of salt. (They may be worse in individual testing)

Regular defensive RAPM
2019 Giannis ranked 2nd behind PG
2020 Giannis ranks 1st by a lot

Regular luck adjusted defensive RAPM
2019 Giannis ranked 1st by a lot
2020 Giannis ranks 1st by a lot

To be clear, ranking second isn’t a sort of death sentence, this is NPI rapm so it’s a bit more noisy and to my knowledge TD never ranked first in his career, although was really close a bunch, Garnett only did so in his Celtics run, and while he was a clear first in 3/5 years 08 was the only year with a large seperation between him and second

Just checking a user named shadows data on apbr who seems respected, giannis’s placements are the same although the gaps are different (he’s a clear second with PG as a crazy first in 2019, and he’s a clear first with a Decent amount of seperation but not a crazy amount over bron in 2020. It’s still roughly equal to most of they Boston KG years though, and it should be noted high end defensive values are higher in general).

To be blunt, I’m way too lazy to make a Google doc with the files from the shotcharts website, especially since it’s easily publically available and accessible. While comparing across seasons with standard deviations is important, for this I’ll only put the top 2 values each year with the name of first place if it seems significant.

While this seems dumb and might not tell enough, I think it’ll be clear what im trying to say when I put in the data. For giannis’s 2020 year I’ll add more incase there’s doubt for that year specifically

RAPM regular 2010
2.77 (Tim Duncan)
2.7
RAPM regular 2011
4.25 (Garnett)
3.19
RAPM regular 2012
4.21 (Taj Gibson)
3.46
RAPM regular 2013
4.06 (Garnett)
3.39
RAPM regular 2014
2.47 (Iggy)
2.21
RAPM regular 2015
2.57 (Tony Allen)
2.55
RAPM regular 2016
2.54 (Kawhi)
2.45 (Green)
RAPM regular 2017
2.32 (Gobert)
2.03
RAPM regular 2018
2.73(Covington)
2.56
RAPM regular 2019
2.03 (Wayoff P)
1.79 (Giannis)
RAPM regular 2020
3.49 (Giannis)
2.49 (Matthews)
2.28 (Schroder)
RAPM regular 2021
2.74 (Gobert)
2.25 (Conley)
2.1 (PJ Dozier)
RAPM regular 2022
2.41 (George Hill)
2.23 (Kenrich Williams)

The overall DRAPM declines as a whole I think after a certain point, Giannis’s is almost certainly the highest here in terms of seperation from second as a percentage and in terms of standard deviations from 0.

This site has luck adjustments too

LARAPM 2010
2.99 (Bogut)
2.87
LARAPM 2011
5.39 (Garnett)
4.65 (Dwight)
3.99 (Caron Butler)
3.89 (Asik)
LARAPM 2012
3.01 (Taj Gibson)
2.87
LARAPM 2013
3.93 (Garnett)
3.14 (Sanders)
2.61
2.59
LARAPM 2014
4.11 (Splitter)
4.09 (Kemba!?)
3.95 (Cp3)
LARAPM 2015
2.47 (Tony Allen)
2.21 (Bogut)
LARAPM 2016
2.33 (Jokic!?)
2.19 (Kawhi)
2.1 Tim Duncan)
LARAPM 2017
1.82 (Gilchrist)
1.81 (Covington)
1.76 (Dray)
LARAPM 2018
2.04 (Gobert)
1.8 (Dejountay)
LARAPM 2019
2.18 (Giannis)
1.71 (Bledsoe)
1.71 (Lopez)
1.69 (Turner)
1.61 (Gobert)
LARAPM 2020
3.19 (Giannis)
2.07 (Matthews)
2.04 (Middleton)
1.93 (Lopez)
1.8 (Marc)
LARAPM 2021
2.11 (BaldEagleOfTruthCaruso)
1.87 (Gobert)
1.66 (Conley)
LARAPM 2022
1.98 (Horford)
1.41 (Draymond)
1.38 (Lonzo)

Added more for giannis since I think a collinearnity issue is something talked about with RAPM stuff, so that’s something worth mentioning here with it being maybe similar to the Warriors RAPM stuff on offense being too dominant during the Curry era to properly assess impact

In any case, the only other high minutes guy in the 2020 squad with a def net rtg of more than -3 is Matthews at -6.2 (99.9 on court def rtg). Then it’s DiVincenzo at -2.7 (101.7) and lopez at -2.3 (102.1). Meanwhile the entire starting lineup outside of giannis is a near neutral or has a positive (bad) def net rtg, George hill has a -6.4 net rtg with an on court rtg of 101.6, but played less than a quarter of the minutes (and it was a trade). Others that ranked well basically played less than 509 minutes, and the starters were all essentially neutrals.

LEBRON data ranks 2020 Giannis as 6th and 2019 Giannis as 30th over the past 15 years, caveat that I don’t think it’s adjusted for standard deviations per season (I think it’s the raw values) and more importantly it tends to overrate bigs, and Giannis doesn’t block as many shots as some others do.

Beyond that, the bucks defense despite Giannis playing limited minutes was genuinely ATG, both first place defence a, 2020 in particular in the games Giannis played was a -8.1 rel DRTG defense, which I think might be the best non Celtics mark ever (slightly worse than 08) although I might be mistake I think it outdoes any of the Spurs marks

(Caveat that a -8 defense wi a 110 avg off rtg isn’t a higher percentage drop relative to the league than a -7 vs a 70 average or something which makes more sense to use honestly, at the same time the current era is one where players rest guys and Bud REALLY rests giannis a lot to be safe so maybe it evens out in terms of pure level of play)

Now if Giannis had a problem defending in the playoffs I get it, but their playoff defense in 2019, 2021, and 2022 was the best in the nba

2019 -7.1 playoff league avg def rtg
2020 -2.8 playoff league avg def rtg
2021 -6.7 playoff league avg def rtg
2022 -7.4 playoff league avg def rtg

They had a problem with dropping too much esp in 2020 iirc (I think they adjusted after giannis got hurt lol) I don’t really blame giannis for coaching incompetence tho, they actually varied their pick and roll coverage more


More importantly to me though, this does make it seem like they were coasting in the RS, I mean at the end of the day 2021 and 2022 bucks were far from a decent defense despite their RS numbers

(From here onward I’m using NBA.com data, just because it didn’t have league averages before but that’s not that important here)

As for his RS defense vs playoff defense

2021 RS
Giannis def net rtg -3.4 (107.4 on court)

2021 playoffs
Giannis def net rtg -5.6 (103.7 on court)

(Second best high minutes guy is Lopez, -0.4 and 105.4, I don’t know if I’m tripping but EVERY rotation player outside of that is a negative in defense in raw impact lol).

Checking it more so, it looks like some of it comes from the heat series, which they just blew them tf out and munched on them in garbage time, altho even taking that out, giannis is now has a def net rtg of -7.5 and an on court def rtg of 105.6, and PJ is next best at -3.2 with an on court of 105.8. Unless I’m tripping everyone other rotation guy is still a negative in defense though

Looking at the teams results they played against a first ranked offense in the nets (before someone says they Kyrie got hurt or harden missed the first part of the series, they were the first ranked offense witth harden and Durant missing more than half the season and Kyrie missing a third), the 18th offense in the heat, the 9th ranked offense in the hawks and the 7th ranked offense in the Suns).

2021

Vs the heat (110.6)
95.4 def rtg
Giannis 95.4 def rtg on-court (+5.4)

Vs the nets (117.3)
107.3 def rtg
Giannis 103.2 def rtg on-court (-11.7)

Vs hawks (114.3)
109.1 def rtg
(107.7 in the games giannis played)
(111.7 in the games giannis missed)
Giannis 106.2 def rtg on-court (-3.6)

Vs Suns (116.3)
112.1 def rtg
Giannis 108.0 def rtg on court (-12.4)

2022
Vs bulls (112.7)
94.4 def rtg
Giannis 91.3 def rtg on court (-4.3)

Vs Celtics (113.6)
108.8 def rtg
Giannis 104.2 def rtg on court (-15.1)

Some of this is mildly misleading, the Suns finished the season strong on offense as did the heat, and the hawks. The nets actually stayed like that consistently, the Celtics had the best offense over the second half of the year while the bulls were pretty poor as the season ended, but let’s just take it at face value

But viewing it as where teams were in terms of their level of play going into the playoffs would make the defense look even more favorable

Nevertheless, the bucks defense over these 6 series was very consistently elite, and as a whole on average probably on the ATG scale, so with giannis impact data looking like it’s the one that looks different I do think it’s fair to say the change was at least partially from him. His tracking data doesn’t change all too much but it’s still similar to his 2019 data if not better I think, although I’d have to check again but it definately wasn’t substantially worse

It’s hard to get definitive proof that he’s a better playoff defender than RS defender. I DONT think he’s quite as good as he was in the 2020 RS at least on a per minute basis (other than the finals lol) but at the same time by all metrics relative to era on a per minute basis that might’ve been one of the most dominant defensive regular seasons since play by play data has been announced, in fact impact stuff it does probably come out to first relative to his peers from the 1999-2021 span, so that’s not saying much

Otoh, I do think that the bucks being a decent RS defense while pretty obviously being a super elite defense in the playoffs, with the main person whose impact data gets better both years is giannis, is pretty significant.

Even this year, only the Warriors did better against the Celtics, and that’s with Tatum forgetting how to play basketball (and this is pretty much entirely due to their defense in the non-giannis minutes, they were a 104.2 offense with giannis in the court which should coincide with their starters, a 119.3 offense with him off the court. For comparison the offense with dray or Wiggins was 103.2 and 103.4).

As a whole, the bucks with giannis on the floor have been a hyper elite defense in 6 series in a row now, and taking the series as a whole (also keep in mind the Celtics and Suns had great second halves of the year and were the best offenses in the league those years from the midway point onwards after a sluggish start, although the Celtics playoff offense was a bit weird to me with Tatum being so inconsistent)

I don’t think it’s outlandish to say Giannis is as good as his 2020 RS in the playoffs, although maybe I wouldn’t go that far, tracking data isn’t full proof at all but it was utterly absurd in 2020 (and as a small note it was suuuper similar to 2016 playoff bron! )

In terms of like, his defensive attributes I don’t understand why this would be an outlandish take either

Giannis is a high IQ defender, rarely makes mistakes, has an absurd motor, and is a guy that can legitimately guard 1-5, 6ft11 with a freakishly long wingspan (that 7ft3 was measured when he was 6ft8.5, hes now at the point that he can casually grab the rim tiptoing now, 7ft6 is probably a safe bet)

Can cover stupid ground, like his max vert might not be absurd but I’ve never seen someone jump as far as he does, which might be more important for his defensive role in 2020 and 2021 where he was breaking up actions as a helpside defender (he’s probably the only guy ever other than wilt that dunks from the ft line because it’s comfortable that way lmao)

We tend to either underestimate the defensive IQ of current defenders or overestimate past defenders, sure Garnett and Duncan have a higher defensive Iq than giannis and guys like AD or Embiid but it’s not as if it’s comparing Magnus Carlson to like a kid playing chess for the first time lol, and draymonds smarter than all of them anyway.

Like generally the elite defenders are elite mentally as well, guys like giannis despite being freakishly athletic aren’t an exception to that, even if he’s not as cerebral as let’s say a draymond green is


https://www.reddit.com/r/MkeBucks/comments/9pq5vv/giannis_casually_grabbing_the_rim_with_his_feet/)

Overall, I think he’s an elite DPOY type defender. He’s shown the ability as a very deserved DPOY and especially 2020 he’s shown ATG defensive ability. His physical profile and motor and overall defensive IQ are in line with that as are his results, and I believe his playoff defense shows that

Like the Gap between him and garnett to me for example is probably mainly that Garnett is better at pick and roll coverages

I can fully understand not being as high on him defensively as I am but him being top 2 in player of the year voting only once and falling out of the top 3 is wild to me. At the very least he’s an ATG defender

I feel offensively there’s not really much to say, his impact data looks a bit worse because of the Miami series (the offense wasn’t bad or anything, but they killed them in garbage time and it’s hard to take much negative in a series that was such an utter sweep like that)

Beyond that if we’re evaluating him as a player by virtue of being a Greek Freak Giannis is kind of an amazing off ball big, from the virtue of setting good screens being stupidly athletic and being an incredible finisher, so he’s highly portable in an off ball role, although jrue isn’t the type that maximizes him in that role so it’s hard even though jrue is great.

Beyond that, imagining Giannis’s defensive impact in a role where he doesn’t have to do as much offensively would be insane considering his defensive impact already is where it’s at when he puts his mind to it, and he’s a good passer off the short roll too

Giannis honestly works REALLY well as a pick and roll partner, and it would probably give him the energy to be a historic level defender as well

On a side note: holding 2021 giannis’s injury against him when he then had potentially one of the absolute greatest finals performances ever on one leg on an injury that took him out of the first few days of training camp 3 months later is MAD corny lmao, that made it 10x more legendary and the fact that they might not win if someone hit the reset button doesn’t matter much to me since 1. He came back stupidly early and killed the city of Phoenix 2. In real life they actually won which really is all that matters vs whatever hypothetical scenario we draw up where him getting hurt leads to them losing.

I’m not 100% on him here, but I do think that his defense seems to be viewed as “oh it’s quite good” sometimes, he’s a legitimate DPOY type player that has shown the ability at least to match absolutely top of the line ATG defensive peaks, and may do so in the playoffs.

Giannis had one of the absolute best two way playoff performances ever

Admittedly, it’s a bit lame that I don’t have like much video watching for this cuz I just don’t have any time, so that’s a bit annoying, either way though giannis for me on defense is a guy that doesn’t play a role that impacts every shot or anything but his motor makes him impact the shots he can to the point that the defense actively does things so he can’t help because he’s such an absurd deterrent


04 KG

- extremely portable, absurd impact data, would be utterly dominant today, but his playoff scoring is an issue for me

[falcolombardi]

On the convo about 2021 being indicitive, their main issue was offensive rebounding and the games Curry missed, their effeciency was solid but I do think you do need multiple decent passers in a motion offense and they didn’t really have that outside of Curry and dray

Anyways, was gonna type this up for the next one but changed it around a bit

2017 Curry

Was thinking of putting Giannis above but I feel it was the perfect offensive performance on a great team, level of play it’s hard to beat but I don’t know if I’d pick Curry over giannis in a vacuum



2021 Giannis

I’m a bit conflicted here, but contextually I do think he should be on here already.

One thing I will say though is that I was looking through the retro player of the years for 2019 and 2020, while I thought he deserved a spot here, my assumption for why Giannis hasn’t been chosen yet or had much consideration was because people were lower on his offense than I was, I now realize it looks like people really didn’t rate him highly on defense?

The peak Duncan vs peak Giannis on offense thread suprised me a bit, I didn’t think it would be so clear cut, my assumption was that people weren’t high on his offense as much as his defense.

The basis of my argument comes from 2021 and 2022 giannis not being a different defender than 2019 or 2020 giannis, rather less effort in the regular season (and I know this past season they mixed up their screen coverages like Miami for postseason versatility), supported by his impact data in the playoffs, the bucks going from league average to probably best or near best in the league defensively in the playoffs

I don’t think 2021 or 2022 Giannis is intrinsically a different defender than his 2019 or 2020 versions, I do think that in general they took those defensive seasons off and ramped it up in the playoffs.

So my arguments on 2021 Giannis defensively will revolve around the impact of his defense in 2019, and in 2020 first and foremost

I wasn’t going to make much of a take about giannis’s defence mostly because I didn’t realize what the general opinion was vs I guess my opinion on his defence.

However, for this to make sense a base assumption is that 2021 playoff giannis on D is the same as 2019 or 2020 RS giannis on D, although there’s an argument for him being better than his 2019 self at least.

So ive kinda gotten the vibe that people (based off retro player of the year voting) didn’t think of him as a super clear DPOY those two years? 2020 I can understand because of ADs postseason exploits, but purely RS wise, those two were imo the best DPOY seasons post 2010.

I realize that might seem like a wild take (I honestly thought it was a pretty normal take) but It’s just that… it just seemed kind of obvious?

First of all, data when he did play vs when he didnt

NOTE: for games with giannis, I got it for 2019 but since it was the same as the on court percentage +_

2019
72 games
105.3 def rtg with giannis

10 games, 108.5 def rtg
(107.7 def rtg 8 games, 2 without starters)


- Avg off rtg faced (10 games) - 108.9
- Avg off rtg faced (8 games) - 108.5

2020
102.5 def rtg with giannis 63 games

108.6 (7 games)
108.5 (10 games, 2 in bubble 1 without starters)

- avg off rtg faced 109.2
- Avg off rtg faced 108.45

So in the 20 or so games that they played without giannis, the bucks were a slightly better than league average defense given their competition, taking only the 15 games their starters played or that weren’t in the bubble, they’re still slightly above average

Not the biggest of samples and there are a few outlier results both ways as is the nature of games in small samples, but still not insignificant. The bucks with giannis otoh, are a -5.2 defense in 2019 and a -7.7 defense in 2020.

Keep in mind budenholzer is weird with rotations, giannis only plays 30 minutes a game not because he’s unable to play longer but because budenholzer is known to be SUPER conservative in this regard

His on court and raw net rtg compared to other ATG defensive teams

https://syndication.bleacherreport.com/amp/2185159-ranking-the-nbas-20-best-defenses-of-all-time.amp.html

On court def rtg

2019 bucks
G 104.5 in 2019 (-3.7 net) (110.4 league avg)

2020 bucks
G 99.6 in 2020 (-7.9 net) (110.6 league avg)

1999 Spurs
David Robinson 92.0 (-2.4) (102.2 league avg)

2004 pistons
Wallace 94.8 (-2.0 net) (102.9 league avg)

2004 Spurs
Duncan 92.1 (-5.4 net) (102.9 league abg)

2005 Spurs
Duncan 94.5 (-8.9 net) (106.1 league avg)

2008 Celtics
Garnett 97.3 (-4.1 net) (107.5 league avg)

2014 pacers
Hibbert 98.9 (-2.9 net) (106.7 league avg)

In this comparison Giannis looks pretty solid, his 2019 season maybe a bit behind, his 2020 season fully deserving to be here.

I’m of the belief that top end defensive impact is limited in the past few years, the advent of three point shooting just means there are less shots to contest at the rim, and the pick and roll means especially in the playoffs you need versatility. Furthermore, even in the RS this leads to the defensive impact one individual can have to be a bit limited in comparison.

Kind of similar to how offensive numbers are inflated nowadays, defensive numbers are probably deflated (in terms of blocked shots and everything) because less blockable shots at the rim occur compared to back then (like it’s easier to block someone driving in a Congested paint than in a 5 out while ur drawn out to the three point line)

I think offensive impact isn’t as inflated maybe because general offensive “scheme evolution” and to a large extent more lax rules as well just means teams can do a bit better when their stars leave the court, but this pretty much has the opposite effect defensively

I think here’s where a bit of an issue comes in for me. Like how I believe it’s probably silly to say, oh devin booker averaged XXX he’s as good as Kobe, it’s also similarly strange for me to think that all the best defenders of today are clearly inferior to the defenders back then.

Or another way to say it, comparing the best defenders now to the best defenders back then, maybe I would get it more had TD and KG been so far head over shoulders above everyone else in defensive impact stuff, the first few Boston KG and some years of TD were, but as a whole while they stand out it’s not in the sense that they consistently dwarfed everyone else like currys offensive impact stuff I think, rather it seems there were usually a few guys that had really high impact in general

Another way to say it, I think the potential defensive impact today has lessened, and that especially on good teams if you transport a guy like Mutombo to todays game, he would not be AS impactful, despite his in eta impact based off rapm and what not making him seem like a (non russell) GOAT defender candidate with how much he dwarfed everyone else

Like, when we talk about how KG is perfect defensively today, I agree, but I don’t think that means he’s more impactful today than he was in the 2000s, I think this means he has less of a dropoff than some others would. This isn’t to say the others wouldn’t be absolutely fantastic but I think defenders have less impact in general today. I feel if you take a Rudy Gobert, whose absurdly dominant as an interior defender he’s probably gonna be a perennial DPOY candidate

Like he’s probably about as good as a mutombo, but even if we say a bit worse, who by some RAPM data consistently dwarfed the competition, and some of his post prime years were competitive or even better than some prime TD years, so I think peak Gobert is probably in the running for DPOY and in that category, even if he probably didn’t peak as high on that end (thats a whole can of worms)

The likes of TD and KG do dominate their era in metrics like defensive RAPM far more than Gobert does, although bigs don’t tend to dominate that as much as they did back then.

I used bball ref for the stuff above, admittedly just because it is far easier and older data seems hard to find.

Luck adjusted RAPM vs regular RAPM is a can of worms imo, luck adjustments generally work better in testing but for multiple years apparently they’re worse so take that with a grain of salt. (They may be worse in individual testing)

Regular defensive RAPM
2019 Giannis ranked 2nd behind PG
2020 Giannis ranks 1st by a lot

Regular luck adjusted defensive RAPM
2019 Giannis ranked 1st by a lot
2020 Giannis ranks 1st by a lot

To be clear, ranking second isn’t a sort of death sentence, this is NPI rapm so it’s a bit more noisy and to my knowledge TD never ranked first in his career, although was really close a bunch, Garnett only did so in his Celtics run, and while he was a clear first in 3/5 years 08 was the only year with a large seperation between him and second

Just checking a user named shadows data on apbr who seems respected, giannis’s placements are the same although the gaps are different (he’s a clear second with PG as a crazy first in 2019, and he’s a clear first with a Decent amount of seperation but not a crazy amount over bron in 2020. It’s still roughly equal to most of they Boston KG years though, and it should be noted high end defensive values are higher in general).

To be blunt, I’m way too lazy to make a Google doc with the files from the shotcharts website, especially since it’s easily publically available and accessible. While comparing across seasons with standard deviations is important, for this I’ll only put the top 2 values each year with the name of first place if it seems significant.

While this seems dumb and might not tell enough, I think it’ll be clear what im trying to say when I put in the data. For giannis’s 2020 year I’ll add more incase there’s doubt for that year specifically

RAPM regular 2010
2.77 (Tim Duncan)
2.7
RAPM regular 2011
4.25 (Garnett)
3.19
RAPM regular 2012
4.21 (Taj Gibson)
3.46
RAPM regular 2013
4.06 (Garnett)
3.39
RAPM regular 2014
2.47 (Iggy)
2.21
RAPM regular 2015
2.57 (Tony Allen)
2.55
RAPM regular 2016
2.54 (Kawhi)
2.45 (Green)
RAPM regular 2017
2.32 (Gobert)
2.03
RAPM regular 2018
2.73(Covington)
2.56
RAPM regular 2019
2.03 (Wayoff P)
1.79 (Giannis)
RAPM regular 2020
3.49 (Giannis)
2.49 (Matthews)
2.28 (Schroder)
RAPM regular 2021
2.74 (Gobert)
2.25 (Conley)
2.1 (PJ Dozier)
RAPM regular 2022
2.41 (George Hill)
2.23 (Kenrich Williams)

The overall DRAPM declines as a whole I think after a certain point, Giannis’s is almost certainly the highest here in terms of seperation from second as a percentage and in terms of standard deviations from 0.

This site has luck adjustments too

LARAPM 2010
2.99 (Bogut)
2.87
LARAPM 2011
5.39 (Garnett)
4.65 (Dwight)
3.99 (Caron Butler)
3.89 (Asik)
LARAPM 2012
3.01 (Taj Gibson)
2.87
LARAPM 2013
3.93 (Garnett)
3.14 (Sanders)
2.61
2.59
LARAPM 2014
4.11 (Splitter)
4.09 (Kemba!?)
3.95 (Cp3)
LARAPM 2015
2.47 (Tony Allen)
2.21 (Bogut)
LARAPM 2016
2.33 (Jokic!?)
2.19 (Kawhi)
2.1 Tim Duncan)
LARAPM 2017
1.82 (Gilchrist)
1.81 (Covington)
1.76 (Dray)
LARAPM 2018
2.04 (Gobert)
1.8 (Dejountay)
LARAPM 2019
2.18 (Giannis)
1.71 (Bledsoe)
1.71 (Lopez)
1.69 (Turner)
1.61 (Gobert)
LARAPM 2020
3.19 (Giannis)
2.07 (Matthews)
2.04 (Middleton)
1.93 (Lopez)
1.8 (Marc)
LARAPM 2021
2.11 (BaldEagleOfTruthCaruso)
1.87 (Gobert)
1.66 (Conley)
LARAPM 2022
1.98 (Horford)
1.41 (Draymond)
1.38 (Lonzo)

Added more for giannis since I think a collinearnity issue is something talked about with RAPM stuff, so that’s something worth mentioning here with it being maybe similar to the Warriors RAPM stuff on offense being too dominant during the Curry era to properly assess impact

In any case, the only other high minutes guy in the 2020 squad with a def net rtg of more than -3 is Matthews at -6.2 (99.9 on court def rtg). Then it’s DiVincenzo at -2.7 (101.7) and lopez at -2.3 (102.1). Meanwhile the entire starting lineup outside of giannis is a near neutral or has a positive (bad) def net rtg, George hill has a -6.4 net rtg with an on court rtg of 101.6, but played less than a quarter of the minutes (and it was a trade). Others that ranked well basically played less than 509 minutes, and the starters were all essentially neutrals.

LEBRON data ranks 2020 Giannis as 6th and 2019 Giannis as 30th over the past 15 years, caveat that I don’t think it’s adjusted for standard deviations per season (I think it’s the raw values) and more importantly it tends to overrate bigs, and Giannis doesn’t block as many shots as some others do.

Beyond that, the bucks defense despite Giannis playing limited minutes was genuinely ATG, both first place defence a, 2020 in particular in the games Giannis played was a -8.1 rel DRTG defense, which I think might be the best non Celtics mark ever (slightly worse than 08) although I might be mistake I think it outdoes any of the Spurs marks

(Caveat that a -8 defense wi a 110 avg off rtg isn’t a higher percentage drop relative to the league than a -7 vs a 70 average or something which makes more sense to use honestly, at the same time the current era is one where players rest guys and Bud REALLY rests giannis a lot to be safe so maybe it evens out in terms of pure level of play)

Now if Giannis had a problem defending in the playoffs I get it, but their playoff defense in 2019, 2021, and 2022 was the best in the nba

2019 -7.1 playoff league avg def rtg
2020 -2.8 playoff league avg def rtg
2021 -6.7 playoff league avg def rtg
2022 -7.4 playoff league avg def rtg

They had a problem with dropping too much esp in 2020 iirc (I think they adjusted after giannis got hurt lol) I don’t really blame giannis for coaching incompetence tho, they actually varied their pick and roll coverage more


More importantly to me though, this does make it seem like they were coasting in the RS, I mean at the end of the day 2021 and 2022 bucks were far from a decent defense despite their RS numbers

(From here onward I’m using NBA.com data, just because it didn’t have league averages before but that’s not that important here)

As for his RS defense vs playoff defense

2021 RS
Giannis def net rtg -3.4 (107.4 on court)

2021 playoffs
Giannis def net rtg -5.6 (103.7 on court)

(Second best high minutes guy is Lopez, -0.4 and 105.4, I don’t know if I’m tripping but EVERY rotation player outside of that is a negative in defense in raw impact lol).

Checking it more so, it looks like some of it comes from the heat series, which they just blew them tf out and munched on them in garbage time, altho even taking that out, giannis is now has a def net rtg of -7.5 and an on court def rtg of 105.6, and PJ is next best at -3.2 with an on court of 105.8. Unless I’m tripping everyone other rotation guy is still a negative in defense though

Looking at the teams results they played against a first ranked offense in the nets (before someone says they Kyrie got hurt or harden missed the first part of the series, they were the first ranked offense witth harden and Durant missing more than half the season and Kyrie missing a third), the 18th offense in the heat, the 9th ranked offense in the hawks and the 7th ranked offense in the Suns).

2021

Vs the heat (110.6)
95.4 def rtg
Giannis 95.4 def rtg on-court (+5.4)

Vs the nets (117.3)
107.3 def rtg
Giannis 103.2 def rtg on-court (-11.7)

Vs hawks (114.3)
109.1 def rtg
(107.7 in the games giannis played)
(111.7 in the games giannis missed)
Giannis 106.2 def rtg on-court (-3.6)

Vs Suns (116.3)
112.1 def rtg
Giannis 108.0 def rtg on court (-12.4)

2022
Vs bulls (112.7)
94.4 def rtg
Giannis 91.3 def rtg on court (-4.3)

Vs Celtics (113.6)
108.8 def rtg
Giannis 104.2 def rtg on court (-15.1)

Some of this is mildly misleading, the Suns finished the season strong on offense as did the heat, and the hawks. The nets actually stayed like that consistently, the Celtics had the best offense over the second half of the year while the bulls were pretty poor as the season ended, but let’s just take it at face value

But viewing it as where teams were in terms of their level of play going into the playoffs would make the defense look even more favorable

Nevertheless, the bucks defense over these 6 series was very consistently elite, and as a whole on average probably on the ATG scale, so with giannis impact data looking like it’s the one that looks different I do think it’s fair to say the change was at least partially from him. His tracking data doesn’t change all too much but it’s still similar to his 2019 data if not better I think, although I’d have to check again but it definately wasn’t substantially worse

It’s hard to get definitive proof that he’s a better playoff defender than RS defender. I DONT think he’s quite as good as he was in the 2020 RS at least on a per minute basis (other than the finals lol) but at the same time by all metrics relative to era on a per minute basis that might’ve been one of the most dominant defensive regular seasons since play by play data has been announced, in fact impact stuff it does probably come out to first relative to his peers from the 1999-2021 span, so that’s not saying much

Otoh, I do think that the bucks being a decent RS defense while pretty obviously being a super elite defense in the playoffs, with the main person whose impact data gets better both years is giannis, is pretty significant.

Even this year, only the Warriors did better against the Celtics, and that’s with Tatum forgetting how to play basketball (and this is pretty much entirely due to their defense in the non-giannis minutes, they were a 104.2 offense with giannis in the court which should coincide with their starters, a 119.3 offense with him off the court. For comparison the offense with dray or Wiggins was 103.2 and 103.4).

As a whole, the bucks with giannis on the floor have been a hyper elite defense in 6 series in a row now, and taking the series as a whole (also keep in mind the Celtics and Suns had great second halves of the year and were the best offenses in the league those years from the midway point onwards after a sluggish start, although the Celtics playoff offense was a bit weird to me with Tatum being so inconsistent)

I don’t think it’s outlandish to say Giannis is as good as his 2020 RS in the playoffs, although maybe I wouldn’t go that far, tracking data isn’t full proof at all but it was utterly absurd in 2020 (and as a small note it was suuuper similar to 2016 playoff bron! )

In terms of like, his defensive attributes I don’t understand why this would be an outlandish take either

Giannis is a high IQ defender, rarely makes mistakes, has an absurd motor, and is a guy that can legitimately guard 1-5, 6ft11 with a freakishly long wingspan (that 7ft3 was measured when he was 6ft8.5, hes now at the point that he can casually grab the rim tiptoing now, 7ft6 is probably a safe bet)

Can cover stupid ground, like his max vert might not be absurd but I’ve never seen someone jump as far as he does, which might be more important for his defensive role in 2020 and 2021 where he was breaking up actions as a helpside defender (he’s probably the only guy ever other than wilt that dunks from the ft line because it’s comfortable that way lmao)

We tend to either underestimate the defensive IQ of current defenders or overestimate past defenders, sure Garnett and Duncan have a higher defensive Iq than giannis and guys like AD or Embiid but it’s not as if it’s comparing Magnus Carlson to like a kid playing chess for the first time lol, and draymonds smarter than all of them anyway.

Like generally the elite defenders are elite mentally as well, guys like giannis despite being freakishly athletic aren’t an exception to that, even if he’s not as cerebral as let’s say a draymond green is


https://www.reddit.com/r/MkeBucks/comments/9pq5vv/giannis_casually_grabbing_the_rim_with_his_feet/)

Overall, I think he’s an elite DPOY type defender. He’s shown the ability as a very deserved DPOY and especially 2020 he’s shown ATG defensive ability. His physical profile and motor and overall defensive IQ are in line with that as are his results, and I believe his playoff defense shows that

Like the Gap between him and garnett to me for example is probably mainly that Garnett is better at pick and roll coverages

I can fully understand not being as high on him defensively as I am but him being top 2 in player of the year voting only once and falling out of the top 3 is wild to me. At the very least he’s an ATG defender

I feel offensively there’s not really much to say, his impact data looks a bit worse because of the Miami series (the offense wasn’t bad or anything, but they killed them in garbage time and it’s hard to take much negative in a series that was such an utter sweep like that)

Beyond that if we’re evaluating him as a player by virtue of being a Greek Freak Giannis is kind of an amazing off ball big, from the virtue of setting good screens being stupidly athletic and being an incredible finisher, so he’s highly portable in an off ball role, although jrue isn’t the type that maximizes him in that role so it’s hard even though jrue is great.

Beyond that, imagining Giannis’s defensive impact in a role where he doesn’t have to do as much offensively would be insane considering his defensive impact already is where it’s at when he puts his mind to it, and he’s a good passer off the short roll too

Giannis honestly works REALLY well as a pick and roll partner, and it would probably give him the energy to be a historic level defender as well

On a side note: holding 2021 giannis’s injury against him when he then had potentially one of the absolute greatest finals performances ever on one leg on an injury that took him out of the first few days of training camp 3 months later is MAD corny lmao, that made it 10x more legendary and the fact that they might not win if someone hit the reset button doesn’t matter much to me since 1. He came back stupidly early and killed the city of Phoenix 2. In real life they actually won which really is all that matters vs whatever hypothetical scenario we draw up where him getting hurt leads to them losing.

I’m not 100% on him here, but I do think that his defense seems to be viewed as “oh it’s quite good” sometimes, he’s a legitimate DPOY type player that has shown the ability at least to match absolutely top of the line ATG defensive peaks, and may do so in the playoffs.

Giannis had one of the absolute best two way playoff performances ever

Admittedly, it’s a bit lame that I don’t have like much video watching for this cuz I just don’t have any time, so that’s a bit annoying, either way though giannis for me on defense is a guy that doesn’t play a role that impacts every shot or anything but his motor makes him impact the shots he can to the point that the defense actively does things so he can’t help because he’s such an absurd deterrent


04 KG

- extremely portable, absurd impact data, would be utterly dominant today, but his playoff scoring is an issue for me

The spurs peaked at a -8.8 relative defense in 2004 regular season which if i am not mistaken is the best defense result since russel (although the 2004 pistons were in a even better pace to end the season when they got rasheed i think, -11 defense in the playoffs which is the best non russel defensive playoffs run)

This is a very good luck at how impactful giannis defense actually is even while sustaining a offense, makes me reconsider giannis higger than i was thinking

[Doctor MJ]

All teams played a bit differently during the 1970s, but some teams really prioritized high post passing offense, Ramsay wasn't the first to do that. Dick Motta's Bulls run a lot of actions in similar way. They focused on off-ball cutting and running around screens, while their center (either Ray or Boerwinkle) run the offense in the high post. Late 1970s Kings teams also tried their version of that. Some other teams didn't abuse it to the same degree, but they used it in a very big part of their offense:

- Bullets with Unseld,
- Suns with Adams,
- Celtics with Cowens.

Of course, there was also one pretty well-known team that did it before the 1970s - 1967-68 Philadelphia 76ers. They relied more on isolation scoring from their perimeter stars, but it was also an offense run through the post from playmaking-first center.

Mostly looking to move on - and very much respect your Mikan vote - but first wanted to thank you for your thoughts here.

I definitely agree that Wilt got used as a playmaking center in the '60s, and all the centers you mention have excellent repuations as passers - Boerwinkle of the Bulls being the guy I think of most here.

The scale of Shaq's talent maybe as much of an outlier or more than Jokic, but he did what he did by taking the basic stuff that was done before, and doing it with a combination of power and agility that was unmatched, rather than doing something that people weren't looking for him to do.

That's not true, the things Shaq did on basketball court were not "the basic stuff that was done before". We've never seen a player playing Shaq's style and dominating the game, even in a fraction of what Shaq did.

I'll just say here:

I remember Shaq as a prospect coming into college. He was a huge, HUGE prospect at the time, recruited by major schools before signing with the coach who had been actively recruiting him since he was 13 years old sending him a letter each week.

I feel like you're talking here based on recognizing fine differences between all these players, but the fact remains, basically all these other guys we're talking about people saw them and said "That looks like a super-star basketball player!" with the possible exception of Bill Russell in high school.

While Jokic's size surely helped him along the way, people didn't think he looked like a basketball super-star even once he was in the NBA.

When the smartest guy is also the least respected prospect, that's telling.

Of course everything we do here is our projection. I think looking at the work certain player did in his era requires less of that than time machine argument though.

So again I'll acknowledge that you are putting your money where your mouth is and voting for Mikan, and to me that makes your perspective seem coherent.

More broadly though, I find the idea of essentially balancing criteria to be pretty problematic.

I can see various options here:

1. Most Significant/Important
2. Most Influential
3. Most Dominant, regardless of era
4. Most Dominant, weighted by respect you have for the talent of the era
5. Most Dominant, going up against the most skilled competition we've witnessed

I'd say you're using 3 whereas typically we've used something we'd call 4 while also using techniques of 5. Obviously, I've reached a watershed where the shape of the game in modern times has made it hard for me to be more era neutral, and I'm still figuring out what that means going forward.

[dooki667]

I don't post much been lurking for years just want to thank y'all for having wonderfully educational discussions and not the typical mud slinging it's a beautiful thing to see

[LA Bird]

Sorry I didn't have time to post the results after the deadline yesterday. There is a tie between 86 Bird and 17 Curry so we have a runoff between the two seasons. If you didn't vote for either of them or you haven't voted in this round at all yet, please do so before 9am ET. If there is still a tie as at the end of the runoff, a winner will be determined according to the tiebreak rule in the project thread.

Spoiler:

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

I will start us off with a vote for 86 Bird. I think you could certainly argue 17 Curry had a slightly better playoffs performance but for the season as a whole, Bird is ahead for me because of a clearly better regular season. If Curry ever had an year where he combined his level of play in the 2016 RS and 2017 PO, then it would be a top 4 peak all time but he never had that strong of a season from start to finish.

[f4p]

1986 Bird. Steph doesn't even have the highest peak of 2017 so it's hard to see how he should be here. I didn't even realize until looking at something else that the Bucks team the Celtics swept in the ECF was +8.7 SRS, so that's pretty impressive. So is being 50-1 at home.

[ardee]

Curry vs Bird is an interesting comparison. These are the MVPs of arguably 2 of the 3 greatest teams of all time. Both have aesthetically pleasing games and aren't known for their athleticism.

I'm going with Bird though. They obviously both can be inconsistent in the Playoffs despite being great in these two years (Curry resolved that once he got stronger but this was before then, and I take his '17 Playoff numbers with a slight pinch of salt given the roster), but Bird was definitely better in the RS in these two years.

I also think Bird was the better defender by a margin (underrated on that end) and while I think Curry has more gravity, Bird's would be a lot more if he played in a more modern era.

Vote: Bird

[Ginoboleee]

Curry vs Bird is an interesting comparison. These are the MVPs of arguably 2 of the 3 greatest teams of all time. Both have aesthetically pleasing games and aren't known for their athleticism.

I'm going with Bird though. They obviously both can be inconsistent in the Playoffs despite being great in these two years (Curry resolved that once he got stronger but this was before then, and I take his '17 Playoff numbers with a slight pinch of salt given the roster), but Bird was definitely better in the RS in these two years.

I also think Bird was the better defender by a margin (underrated on that end) and while I think Curry has more gravity, Bird's would be a lot more if he played in a more modern era.

Vote: Bird

I entirely agree with Ardee's assessment.
Hopefully that is sufficient explanation for my vote to count.
I too vote Bird.

[LA Bird]

Here are the results for round 9

Winner: 86 Bird

There were 21 voters in this round: SickMother, iggymcfrack, DraymondGold, f4p, capfan33, Doctor MJ, Dutchball97, CharityStripe34, ceoofkobefans, trex_8063, trelos6, mdonnelly1989, 70sFan, letskissbro, ardee, Djoker, Lou Fan, jalengreen, Proxy, Dr Positivity, MyUniBroDavis

A total of 40 seasons received at least 1 vote: 03 Garnett, 04 Garnett, 06 Bryant, 08 Bryant, 15 Curry, 16 Curry, 17 Curry, 17 Leonard, 18 Curry, 19 Curry, 20 Antetokounmpo, 20 Jokic, 21 Antetokounmpo, 21 Curry, 21 Jokic, 22 Antetokounmpo, 22 Curry, 22 Jokic, 50 Mikan, 51 Mikan, 58 Pettit, 59 Pettit, 62 Pettit, 65 West, 66 West, 69 West, 76 Erving, 83 Malone, 84 Bird, 85 Bird, 85 Johnson, 86 Bird, 87 Bird, 87 Johnson, 88 Johnson, 89 Johnson, 90 Johnson, 94 Robinson, 95 Robinson, 96 Robinson

Top 5 seasons
17 Curry: 1.000 (38-0)
86 Bird: 1.000 (38-0)
04 Garnett: 0.949 (37-2), loses to 17 Curry, 86 Bird
87 Johnson: 0.923 (36-3), loses to 04 Garnett, 17 Curry, 86 Bird
16 Curry: 0.897 (35-4), loses to 04 Garnett, 17 Curry, 86 Bird, 87 Johnson

H2H record
17 Curry vs 86 Bird: 8-8
17 Curry vs 04 Garnett: 10-4
17 Curry vs 87 Johnson: 8-7
17 Curry vs 16 Curry: 10-1
86 Bird vs 04 Garnett: 11-7
86 Bird vs 87 Johnson: 12-5
86 Bird vs 16 Curry: 10-4
04 Garnett vs 87 Johnson: 10-7
04 Garnett vs 16 Curry: 8-6
87 Johnson vs 16 Curry: 7-4

Tiebreaker
Bird: 3 (LA Bird, f4p, Ginoboleee)
Curry: 0

[trex_8063]

I missed the tie-break vote deadline, but would have voted for Bird. I think peak Curry is the better scorer, and I think his off-ball gravity and movement provides more intangible value than does Bird's [arguably even in a modern-era context].

However, it's the play-making and rebounding edges [+/- a general defensive edge too] that puts Bird slightly over the top for me. Very close though: I have Bird around 12th-14th for peak, Curry at 15th. So it's practically splittin' hairs.

[falcolombardi]

And this would be my reminder as we've now said that 6 of the 8 top players in history play a position that is now largely manned by role players, that in the cutting edge of the game, it's been guys who are a threat on the perimeter that have come to reign supreme.

We are fresh off a season when top 3 MVP candidates played at center, just to remind you

Giannis mostly doesn't play center.

Jokic plays virtually nothing like any of the guys who have been voted in.

Embiid is the closest thing to these guys but with an ability to shoot from 3 and attack from the perimeter, and by the end of the playoffs the question was whether he was Top 5 or not.

Embiid was injured kn the playoffs tho, before them he was unquestionably top 5

Giannis and jokic dont play like most bigs but neither does curry play like most guards and nobody says he doesnt count as one

[Doctor MJ]

We are fresh off a season when top 3 MVP candidates played at center, just to remind you

Giannis mostly doesn't play center.

Jokic plays virtually nothing like any of the guys who have been voted in.

Embiid is the closest thing to these guys but with an ability to shoot from 3 and attack from the perimeter, and by the end of the playoffs the question was whether he was Top 5 or not.

Embiid was injured kn the playoffs tho, before them he was unquestionably top 5

Giannis and jokic dont play like most bigs but neither does curry play like most guards and nobody says he doesnt count as one

1. Right, but the feeling for many is that he got surpassed in the playoffs, not simply that he played worse.

2. There's not a question of whether guards can thrive today the way there is with bigs.

[Black Feet]

We are fresh off a season when top 3 MVP candidates played at center, just to remind you

Giannis mostly doesn't play center.

Jokic plays virtually nothing like any of the guys who have been voted in.

Embiid is the closest thing to these guys but with an ability to shoot from 3 and attack from the perimeter, and by the end of the playoffs the question was whether he was Top 5 or not.

Embiid was injured kn the playoffs tho, before them he was unquestionably top 5

Giannis and jokic dont play like most bigs but neither does curry play like most guards and nobody says he doesnt count as one

Embiid was second in MVP votes, no argument for him outside the top 5. Especially having to play with no Simmons the whole year and then Harden who played like crap.

[falcolombardi]

Giannis mostly doesn't play center.

Jokic plays virtually nothing like any of the guys who have been voted in.

Embiid is the closest thing to these guys but with an ability to shoot from 3 and attack from the perimeter, and by the end of the playoffs the question was whether he was Top 5 or not.

Embiid was injured kn the playoffs tho, before them he was unquestionably top 5

Giannis and jokic dont play like most bigs but neither does curry play like most guards and nobody says he doesnt count as one

1. Right, but the feeling for many is that he got surpassed in the playoffs, not simply that he played worse.

2. There's not a question of whether guards can thrive today the way there is with bigs.

If he didnt get injured maybe sixers beat miami and nobody ranks butler above embiid, sixers went 2-2 with a hurt joel playing

There are 3 bigs playing at a higher level than any non curry-guard (and jokic and giannis prolly better than curry too) right now

If we ranked curry top 5 in 2021 i dont see no reason to not have embiid top 5 in 2022, injured playoffs shouldnt drop off a player more than no playoffs
(i have 2022 embiid as close to 2021 curry in regular season impact)