2021-22 NBA Season Discussion

[Doctor MJ]

I think it's hard to point to any stat other than Raw +/- that would put Tatum on Giannis level, and the difference between a +10 and +14 player seems vulnerable to noise in that regard.

Hmm, so let's get clear on a few things here:

Raw +/- is your total +/- when you're on the court. Here's how Tatum & Giannis look there:

Tatum +607
Giannis +361

This is a massive, massive difference, and all the more amazing given that their two teams have comparable records.

From there, it certainly makes sense to ask whether Tatum's edge holds up with a more sophisticated version of the stat like RAPM, and the answer is yes. Tatum leads the league in RAPM by nbashotcharts with a massive lead over Giannis.

So this is not a situation where where I'm simply pointing at On/Off and relying a lot on Off numbers (which are of course prone to small sample size concerns) to make some statement here. Tatum does look better based on this sort of data than Giannis this year by regression, and in terms of what it's meant On court over the course of the year, it's meant a lot.

To put it another way:

If these players were on team records that were in proportion to their On court numbers, Tatum would be on a team that was drastically higher than Giannis, and he WOULD have been getting MVP attention a lot time ago. What's propping Giannis up is how well the Bucks now do without him. That's only reason they are seen as "eh, similar team accomplishment" in the first place, which then leads people to say "And we know Giannis is better so, he must be getting less help than Tatum", but the data tells a clearly different story.

Doesn't mean Tatum's capable of more with everything on the line against the top competition in the playoffs.
Doesn't mean Tatum's been more valuable this season by these metrics than Giannis was when he was winning MVPs.

But if we're judging these players by actual impact this season, rather than by some combination of team record, see, box score production, and pre-existing assessment, it should not be easy for anyone to dismiss Tatum > Giannis.

Tatum has played like 30% more minutes than Giannis, so once that is taken into account the total +/- number you posted does not paint that much different a picture than their on/off of +14 vs +10. If someone wants to take a heavy games/minutes played viewpoint to argue Tatum>Giannis, I guess that's reasonable, but in terms of the games played impact I can't say i'm going to take it a lot into account that Tatum's +/- is better than Giannis when both are good in that area. Tatum's +/- is mostly better cause the Celtics are 4 points worse on offense when he doesn't play than the Bucks without Giannis (Bucks offense with Giannis is slightly better than Celtics with Tatum). That could have something to do with their play but it seems plausible to me that some teams are better made up to play without their star scorer more than others. In regards to nbashotchart, one season RAPM has always had issues, for example George Hill is currently first on their DRAPM, and Whiteside is 3rd on ORAPM. Tatum is 9th on DRAPM and Giannis is 97th, which I'm not sure is justified. RPM has the same issue where Tatum's DRPM is much higher than Giannis.

Re: Tatum's played a lot more. Doesn't really hurt the Tatum argument does it?

Re: In games played...Tatum still has the edge in OnCourt, On/Off, and RAPM.

Re: Tatum's mostly better because the team hurts more on offense when he's not in whereas the Bucks do pretty well. Indeed. Which is all the more reason why it's amazing the Bucks don't have a better record. People are getting fooled because the Bucks still have a shot at the #1 seed, but they are performing VASTLY worse than they did when Giannis won MVPs despite the fact that their 2nd best player now (Jrue) wasn't even on the team back then.

Re: one season RAPM has issues. Ah, let's be clear: When we say stuff like this, what we've always meant is that the regression can have noise to it. But in this case there's no reason to think that's an issue because the raw data is so clear cut.

(And let's note, since you're skeptical about one year RAPM, the same site also has a 3 and a 5 year model. In the 3-year model Tatum ranks 2nd in the league, in the 5-year, he ranks 5th. We're not actually talking about that much of anomaly right now - we're talking of a continued growth above something that was already quite significant but many were not noticing.)

On the other hand, I'm certainly not saying that this RAPM means that Giannis is literally a much worse player than he used to be in the sense that he's no longer capable of doing what he did before. What I'd say we're seeing from Giannis is the same thing we see from most veterans: Once the regular season offers them nothing more to prove, they take it less seriously even if they still make sure to get their stats.

And Giannis has been pretty clear about this when asked about concerns about this regular season:

I find the +/- evidence more convincing if it's backed up logically why someone has such a high impact beyond boxscore, eg. Draymond Green's makes sense why the numbers wouldn't capture it. In the case of Tatum I'm not sure what he's doing that's supposed to make him such a non boxscore king unless people think he's the main driving force for their defense, which I don't think Celtics fans do.

So, I think you've learned the wrong lessons about box score & +/- then. It's not a thing where the box score captures everything from classical stars and is only useful to identify other players who also have impact. The reality is that the box score always misses a ton of what's happening out on the court, and so when we see players with outlier +/- numbers over considerable sample, it always makes sense to ask what's happening.

In terms of what Tatum's doing out there, I think in one way the answer is simple:

You're talking about a guy who's always performing well by these metrics on both sides of the ball. He's a volume scorer scoring on solid efficiency while being a perimeter guy - which basically always means he's adding extra value compared to a similarly stated big man on offense - and defensively he's very solid there too.

I say all this but to be clear: I didn't predict Tatum would stand out to the degree he is this season, and I've never been particularly enthralled with him and his scraggly facial hair. Nevertheless, the data is eye-opening and not to be dismissed lightly.

[Doctor MJ]

Using raw plus minus against Giannis given that he's specifically tasked with being shoddy lineup glue for much of his minutes is misguided at best. Particularly this season when he's been asked to be big and small at the same time for a lot minutes.

As I've said, it's not just the raw stuff that tells us this. I use the raw stuff because I think everyone can understand it if they want to.

[Doctor MJ]

Harden on the Sixers is making 5.4 FGs per game and 8.7 FTs. He is making more FTs than FGs per game. Pretty interesting.

Another weird free-throw stat about Chris Paul

When team is in non-penalty he averages 1 FTM/75 possessions.

When his team is in the penalty he averages 13.9 FTM/75 possessions.

I'll tell you: This really bothers me about Paul. Not saying Harden doesn't also bother me, because he's the worst, but literally the reason why old man Paul is relevant at this point is largely because of his ability to get the refs to call BS fouls for him right when BS foul calls are the most damaging to the game of basketball. They should be better than this given that it's all so predictable.

[falcolombardi]

Harden on the Sixers is making 5.4 FGs per game and 8.7 FTs. He is making more FTs than FGs per game. Pretty interesting.

Another weird free-throw stat about Chris Paul

When team is in non-penalty he averages 1 FTM/75 possessions.

When his team is in the penalty he averages 13.9 FTM/75 possessions.

I'll tell you: This really bothers me about Paul. Not saying Harden doesn't also bother me, because he's the worst, but literally the reason why old man Paul is relevant at this point is largely because of his ability to get the refs to call BS fouls for him right when BS foul calls are the most damaging to the game of basketball. They should be better than this given that it's all so predictable.

part of it obviously is foul fishing

but i wonder if part of it is "saving" the more agressive play for when the other team is in the bonus cause tactically it makes sense

i thinl paul would be Fine with foul baiting 40 minutes long if it made his team win so i dont think is -only- thst

remember tjis is a guy with a 10% shot profile in the paint i wonder if he saves energy for when the other team is in the bonus and starts attacking the paint

[Doctor MJ]

Harden on the Sixers is making 5.4 FGs per game and 8.7 FTs. He is making more FTs than FGs per game. Pretty interesting.

It is until you realize for his career he averages 7.4 FG/G and 7.5 FT/G

Both stats are interesting, but the fact that the ratio is even more skewed toward BS now in a year where the NBA is supposed to be cracking down is a problem.

For many years I just shrugged at the way players drew BS fouls, but in recent years they've gotten it down to such a science that the NBA needs to completely change its approach. Looked like they might have done this at the start of this season, but clearly a new "emphasis" isn't the same thing as a new philosophy, and a new philosophy is what is required.

[70sFan]

Why can't we just come back to 1970s officiating?

Seriously though, the league has been searching for the solution of the problem they created. Current version of Harden would never exist in the 1970s, because refs were far less willing to give superstars bullsh*t calls and they protected defenders more.

[Doctor MJ]

Grizzlies without ja:
Opponents: 29.9% from 3
Team: 38.3% drom 3
8.4 difference

With ja
Opponnents: 36.6% from 3
Team: 33.7% from 3
-2.9 difference

Total difference: 11.3

Shooting variance really in effect without Ja.

This is a good thing to take a look at and be cautious about drawing big conclusions from, but I'll say I don't think it's a given this goes away with more sample. It's certainly not the case that the Grizz have anyone else who can play the Ja role as well as Ja, but when an offense is built around one guy they play one way, and when that guy isn't there, they necessarily have to change how they play. If they find a different way that works, it can be better than the original way.

As I say this, the main thing I'm thinking is that in the playoffs, I'd guess that the non-Ja offense is a lot less resilient than the Ja offense. Were I the Grizz, I'd still be looking to build that offense around Ja going forward.

[Doctor MJ]

Another weird free-throw stat about Chris Paul

When team is in non-penalty he averages 1 FTM/75 possessions.

When his team is in the penalty he averages 13.9 FTM/75 possessions.

I'll tell you: This really bothers me about Paul. Not saying Harden doesn't also bother me, because he's the worst, but literally the reason why old man Paul is relevant at this point is largely because of his ability to get the refs to call BS fouls for him right when BS foul calls are the most damaging to the game of basketball. They should be better than this given that it's all so predictable.

part of it obviously is foul fishing

but i wonder if part of it is "saving" the more agressive play for when the other team is in the bonus cause tactically it makes sense

i thinl paul would be Fine with foul baiting 40 minutes long if it made his team win so i dont think is -only- thst

remember tjis is a guy with a 10% shot profile in the paint i wonder if he saves energy for when the other team is in the bonus and starts attacking the paint

I think we can combine those two points if we simply acknowledge that Paul being "aggressive" at this point means deciding when to hurl his body at people trying to play basketball based on when it will get him to the line.

[Dr Positivity]

Re: Tatum's mostly better because the team hurts more on offense when he's not in whereas the Bucks do pretty well. Indeed. Which is all the more reason why it's amazing the Bucks don't have a better record. People are getting fooled because the Bucks still have a shot at the #1 seed, but they are performing VASTLY worse than they did when Giannis won MVPs despite the fact that their 2nd best player now (Jrue) wasn't even on the team back then.

I'd say the biggest reason for Bucks decline is health and depth. Giannis/Middleton/Holiday numbers together are pretty good.

I'm not positive the Bucks supporting cast should be rated above the Celtics when health is considered. The Bucks have the 3rd all star name, but it seems like Celtics role players delivered more than the Bucks this year especially on the defensive end.

I typed a longer response but honestly you're just higher on raw +/- than me, my take on its issues are largely covered in my response to Colbinii. As I said the key factor to me is that Giannis +/- and single season RAPM is still very good this year and typical for a superstar season.

[falcolombardi]

Hmm, so let's get clear on a few things here:

Raw +/- is your total +/- when you're on the court. Here's how Tatum & Giannis look there:

Tatum +607
Giannis +361

This is a massive, massive difference, and all the more amazing given that their two teams have comparable records.

From there, it certainly makes sense to ask whether Tatum's edge holds up with a more sophisticated version of the stat like RAPM, and the answer is yes. Tatum leads the league in RAPM by nbashotcharts with a massive lead over Giannis.

So this is not a situation where where I'm simply pointing at On/Off and relying a lot on Off numbers (which are of course prone to small sample size concerns) to make some statement here. Tatum does look better based on this sort of data than Giannis this year by regression, and in terms of what it's meant On court over the course of the year, it's meant a lot.

To put it another way:

If these players were on team records that were in proportion to their On court numbers, Tatum would be on a team that was drastically higher than Giannis, and he WOULD have been getting MVP attention a lot time ago. What's propping Giannis up is how well the Bucks now do without him. That's only reason they are seen as "eh, similar team accomplishment" in the first place, which then leads people to say "And we know Giannis is better so, he must be getting less help than Tatum", but the data tells a clearly different story.

Doesn't mean Tatum's capable of more with everything on the line against the top competition in the playoffs.
Doesn't mean Tatum's been more valuable this season by these metrics than Giannis was when he was winning MVPs.

But if we're judging these players by actual impact this season, rather than by some combination of team record, see, box score production, and pre-existing assessment, it should not be easy for anyone to dismiss Tatum > Giannis.

Tatum has played like 30% more minutes than Giannis, so once that is taken into account the total +/- number you posted does not paint that much different a picture than their on/off of +14 vs +10. If someone wants to take a heavy games/minutes played viewpoint to argue Tatum>Giannis, I guess that's reasonable, but in terms of the games played impact I can't say i'm going to take it a lot into account that Tatum's +/- is better than Giannis when both are good in that area. Tatum's +/- is mostly better cause the Celtics are 4 points worse on offense when he doesn't play than the Bucks without Giannis (Bucks offense with Giannis is slightly better than Celtics with Tatum). That could have something to do with their play but it seems plausible to me that some teams are better made up to play without their star scorer more than others. In regards to nbashotchart, one season RAPM has always had issues, for example George Hill is currently first on their DRAPM, and Whiteside is 3rd on ORAPM. Tatum is 9th on DRAPM and Giannis is 97th, which I'm not sure is justified. RPM has the same issue where Tatum's DRPM is much higher than Giannis.

Re: Tatum's played a lot more. Doesn't really hurt the Tatum argument does it?

Re: In games played...Tatum still has the edge in OnCourt, On/Off, and RAPM.

Re: Tatum's mostly better because the team hurts more on offense when he's not in whereas the Bucks do pretty well. Indeed. Which is all the more reason why it's amazing the Bucks don't have a better record. People are getting fooled because the Bucks still have a shot at the #1 seed, but they are performing VASTLY worse than they did when Giannis won MVPs despite the fact that their 2nd best player now (Jrue) wasn't even on the team back then.

Re: one season RAPM has issues. Ah, let's be clear: When we say stuff like this, what we've always meant is that the regression can have noise to it. But in this case there's no reason to think that's an issue because the raw data is so clear cut.

(And let's note, since you're skeptical about one year RAPM, the same site also has a 3 and a 5 year model. In the 3-year model Tatum ranks 2nd in the league, in the 5-year, he ranks 5th. We're not actually talking about that much of anomaly right now - we're talking of a continued growth above something that was already quite significant but many were not noticing.)

On the other hand, I'm certainly not saying that this RAPM means that Giannis is literally a much worse player than he used to be in the sense that he's no longer capable of doing what he did before. What I'd say we're seeing from Giannis is the same thing we see from most veterans: Once the regular season offers them nothing more to prove, they take it less seriously even if they still make sure to get their stats.

And Giannis has been pretty clear about this when asked about concerns about this regular season:

I find the +/- evidence more convincing if it's backed up logically why someone has such a high impact beyond boxscore, eg. Draymond Green's makes sense why the numbers wouldn't capture it. In the case of Tatum I'm not sure what he's doing that's supposed to make him such a non boxscore king unless people think he's the main driving force for their defense, which I don't think Celtics fans do.

So, I think you've learned the wrong lessons about box score & +/- then. It's not a thing where the box score captures everything from classical stars and is only useful to identify other players who also have impact. The reality is that the box score always misses a ton of what's happening out on the court, and so when we see players with outlier +/- numbers over considerable sample, it always makes sense to ask what's happening.

In terms of what Tatum's doing out there, I think in one way the answer is simple:

You're talking about a guy who's always performing well by these metrics on both sides of the ball. He's a volume scorer scoring on solid efficiency while being a perimeter guy - which basically always means he's adding extra value compared to a similarly stated big man on offense - and defensively he's very solid there too.

I say all this but to be clear: I didn't predict Tatum would stand out to the degree he is this season, and I've never been particularly enthralled with him and his scraggly facial hair. Nevertheless, the data is eye-opening and not to be dismissed lightly.

i dont t know if "giannis is not trying as hard" or "his virtually idéntical stats are somehow much less valuable now" is my conclusión

is 30/10/8 in great efficiency by 2022 -somehow- much less impactful thsn 30/10/8 in great efficiency of 2020? bucks offense is actually much -better- now thsn in 2020 too (of course with better talent) so is not like offense is the "issue" in giannis impact stats

so i dont buy the notion that giannis numbers are less valuable than before without a explanation for why that would be the case

the most plausible explanation is defense (where bucks are mediocre this regular season) is giannis coasting in defense? honestly is almost the opposite to me

he was asked for most of the season to play out og his position to cover for brook López absence, if anythingh it seems like he was asked to carry a bigger defensive load than usual because bucks lack of a secondary center

if tatum lost Robert Williams for most of the season and was asked to defendw out of his natural position/role and without a strong rim protector to help behind i have a hard time imagining hr would look as good as he has in defensive impact numbers

and i have almost zero doubts that giannis with brook López wouldnt have better defense numbers

[Texas Chuck]

I mean part of those Paul numbers are obviously he has the ball in his hands when the other team is forced to intentionally foul in the 4th quarter--something that happens a bunch with Phoenix because they basically always have the lead.

And yes, intelligent players like Paul are going to go to their bag of moves to get themselves to the FT line when there is real advantage to be gained from. I think the answer isn't to be angry at Paul for playing smart, but question why other offensive players aren't doing similar things.

Paul isn't here to play theoretically beautiful basketball. He's here to win. And its pretty hard to argue much against his approach right now.

[falcolombardi]

I mean part of those Paul numbers are obviously he has the ball in his hands when the other team is forced to intentionally foul in the 4th quarter--something that happens a bunch with Phoenix because they basically always have the lead.

And yes, intelligent players like Paul are going to go to their bag of moves to get themselves to the FT line when there is real advantage to be gained from. I think the answer isn't to be angry at Paul for playing smart, but question why other offensive players aren't doing similar things.

Paul isn't here to play theoretically beautiful basketball. He's here to win. And its pretty hard to argue much against his approach right now.

unfortunately we all have a tendency to value more highly players we like more stylistically and viceversa, is hard to not have some amount of bias

so players who play a more aesthetical/liked style of basketball will receive less criticism/scrutiny and those who play "uglier" basketball will be over scrutinized

[MartinToVaught]

Harden on the Sixers is making 5.4 FGs per game and 8.7 FTs. He is making more FTs than FGs per game. Pretty interesting.

Another weird free-throw stat about Chris Paul

When team is in non-penalty he averages 1 FTM/75 possessions.

When his team is in the penalty he averages 13.9 FTM/75 possessions.

This is just disgraceful. The modern level of flopping from Harden, CP3, Embiid, Trae, etc. is every bit as awful in my book as Spygate was in the NFL or the Astros' sign stealing was in baseball. It's basically cheating, and ruins the integrity of the sport.

[Texas Chuck]

unfortunately we all have a tendency to value more highly players we like more stylistically and viceversa, is hard to not have some amount of bias

so players who play a more aesthetical/liked style of basketball will receive less criticism/scrutiny and those who play "uglier" basketball will be over scrutinized

Oh no doubt. And Doc specifically is open that he approaches all of this differently and the theory matters as much or more to him as the actual play.

I try really hard though to fight against my own personal biases in regards to style of play though and accept that impact comes in all kinds of forms regardless of position. You can be Steve Nash or Jason Kidd and have tremendous impact as a PG playing very differently. Or KG or Dirk as PF/C's.

Sometimes I fail admittedly and I have to work especially hard with high scoring iso wings which is my least favorite player archetype. I know I can be guilty of underrating Mike and Wade and Kobe and TMac and Vince etc.. because of it.

But as to Paul, I think he has relatively very few plays that I would call outside the spirit of the game and very many plays where his knowledge of the rules and how the game is officiated allows him to help his team in ways other players miss. I view it as very intelligent and to the positive side of his ledger. Just like when Jason Kidd understood if a coach is standing on the court barking instructions and I can run over there and run into him, he's going to be penalized and my team will benefit. It's not a normal basketball play, but its also smart. I try and never penalize players for being smarter than everyone else. And Paul is smarter than everyone else pretty much every time he's on a court.

[Doctor MJ]

Re: Tatum's mostly better because the team hurts more on offense when he's not in whereas the Bucks do pretty well. Indeed. Which is all the more reason why it's amazing the Bucks don't have a better record. People are getting fooled because the Bucks still have a shot at the #1 seed, but they are performing VASTLY worse than they did when Giannis won MVPs despite the fact that their 2nd best player now (Jrue) wasn't even on the team back then.

I'd say the biggest reason for Bucks decline is health and depth. Giannis/Middleton/Holiday numbers together are pretty good.

I'm not positive the Bucks supporting cast should be rated above the Celtics when health is considered. The Bucks have the 3rd all star name, but it seems like Celtics role players delivered more than the Bucks this year especially on the defensive end.

I typed a longer response but honestly you're just higher on raw +/- than me, my take on its issues are largely covered in my response to Colbinii. As I said the key factor to me is that Giannis +/- and single season RAPM is still very good this year and typical for a superstar season.

A decline in depth would imply that what was happening was that the team was falling apart without Giannis - like what we see in Denver with Jokic - but we don't see that. I'm not saying there isn't a loss of depth, but it's not affecting the Bucks in the way we'd typically expect depth to affect a team with an MVP level player.

Re: health. There's certainly something to that. I expect the Bucks would look a lot better if they had their best version of Brook Lopez.

Re: 3 all-star names doesn't mean a better supporting cast. That's true, on the other hand those other two stars led the team to victory in the 2nd round of the playoffs without Giannis, and there's evidence this season as well that they can lead the offense pretty dang well without Giannis.

Re: Celtics role players delivery more this year especially on defense. Can't disagree with that, but I'd again emphasize that the Celtics +/- trends don't look like a team that's winning based on depth instead of stars. This isn't a situation where like, say, the Melo Nuggets where the team seemed to do okay without him until he left...and then they looked even better. For all the performance of the other players, Tatum's numbers stand out head and shoulders above everyone else - and frankly the same as been true for years now.

For perspective, here are the raw +/- totals of the Celtics over the past 4 years all together:

Tatum +1681
Smart +983
Brown +753

This is a just plain a huge lead. I don't know if you'll find any lead like from any other team among a core that's been together the whole time.

Re: just higher than you on +/-. Yup, and I should probably let you be.

[Doctor MJ]

i dont t know if "giannis is not trying as hard" or "his virtually idéntical stats are somehow much less valuable now" is my conclusión

Whoa. Okay, your phrasing right here shows a point where we strongly diverge in perspective.

I think one of the most important conclusions I've ever drawn is that there's a tendency to assume that production has a consistent proportionality to impact for a given player causes us to put the cart before the horse.

You mention in your post that if Lopez were there the defense would be better, and if he was fully himself, we certainly agree on this...but if the difference between this defensive level (110.4 DRtg with Giannis on the floor) and the '19-20 level (99.6 DRtg with Giannis on the floor) is just about Lopez, then that's attributing way more defensive value to Lopez than to Giannis.

I think that undersells the value Giannis provided to that team, to put it mildly. I think he was incredibly deserving of that DPOY he won.

the most plausible explanation is defense (where bucks are mediocre this regular season) is giannis coasting in defense? honestly is almost the opposite to me

he was asked for most of the season to play out og his position to cover for brook López absence, if anythingh it seems like he was asked to carry a bigger defensive load than usual because bucks lack of a secondary center

if tatum lost Robert Williams for most of the season and was asked to defendw out of his natural position/role and without a strong rim protector to help behind i have a hard time imagining hr would look as good as he has in defensive impact numbers

and i have almost zero doubts that giannis with brook López wouldnt have better defense numbers

So, the whole "out of his natural position" argument is an interesting one to use here.

If I take a DPOY-level center, and make him play point guard, and as a result show less signs of defensive impact, it would certainly be unfair to criticize his effort, or to impugn what he's capable of. To the extent I've done this, I should walk it back. I really don't mean to come off as all that critical of Giannis.

However, from an impact perspective, that center probably shouldn't be getting the DPOY award that year, right?

[Doctor MJ]

Why can't we just come back to 1970s officiating?

Seriously though, the league has been searching for the solution of the problem they created. Current version of Harden would never exist in the 1970s, because refs were far less willing to give superstars bullsh*t calls and they protected defenders more.

Great insight but I'd object to one thing:

I think "created" is too proactive of a word to use. To me what's happened is a bit more like erosion. I think the offensive players here, because they drive the action like they do, have been able to probe and probe and probe, and wherever they find something that's actually challenging for refs to get right, they've gradually made the refs feel like they can't not reward the offensive players unless they see precisely that the offensive player manufactured the contact through non-basketball action.

So I'm pretty forgiving that they didn't stop this from happening, but now they need to fix it, and that's probably going to require a philosophical turn where refs are backed whenever they conclude a player known for flopping is probably flopping, and don't give him the call.

[VanWest82]

Just to chime in and attempt to add some context, I think what might be happening re Tatum is teams have started to give him a version of the MJ/Kobe treatment. It's obviously not as extreme per double team data (https://nbacourtoptix.nba.com/en) but it makes sense when you consider how much less attention the next guy (Brown) gets in comparison vs. Jrue and Khris.

So Tatum becomes key to the Celtics offense even though his assist numbers don't paint the picture of an indispensable offensive player the way Trae or Luka's pts + ast might.

And on defense, even though he might only be the 3rd or 4th most important defensive player in terms of role, we now understand just how important a competent, lengthy, decent IQ forward is to NBA defenses, and most teams (including Celts) don't have so many of those guys that they can just always have one out there.

I don't think Tatum is as good as his +/- based metrics suggest but he's almost assuredly more impactful than his box suggests.

[Doctor MJ]

I mean part of those Paul numbers are obviously he has the ball in his hands when the other team is forced to intentionally foul in the 4th quarter--something that happens a bunch with Phoenix because they basically always have the lead.

And yes, intelligent players like Paul are going to go to their bag of moves to get themselves to the FT line when there is real advantage to be gained from. I think the answer isn't to be angry at Paul for playing smart, but question why other offensive players aren't doing similar things.

Paul isn't here to play theoretically beautiful basketball. He's here to win. And its pretty hard to argue much against his approach right now.

Great, great point to emphasize. I've said it to others many times, and it's still true.

My criticisms of Paul's approach don't affect his value. This is me talking on a aesthetic level about how the NBA needs to manage their product...and getting a bit fanatical/cantankerous about it. My apologies for my tone.

[eminence]

Re: Tatum's mostly better because the team hurts more on offense when he's not in whereas the Bucks do pretty well. Indeed. Which is all the more reason why it's amazing the Bucks don't have a better record. People are getting fooled because the Bucks still have a shot at the #1 seed, but they are performing VASTLY worse than they did when Giannis won MVPs despite the fact that their 2nd best player now (Jrue) wasn't even on the team back then.

I'd say the biggest reason for Bucks decline is health and depth. Giannis/Middleton/Holiday numbers together are pretty good.

I'm not positive the Bucks supporting cast should be rated above the Celtics when health is considered. The Bucks have the 3rd all star name, but it seems like Celtics role players delivered more than the Bucks this year especially on the defensive end.

I typed a longer response but honestly you're just higher on raw +/- than me, my take on its issues are largely covered in my response to Colbinii. As I said the key factor to me is that Giannis +/- and single season RAPM is still very good this year and typical for a superstar season.

A decline in depth would imply that what was happening was that the team was falling apart without Giannis - like what we see in Denver with Jokic - but we don't see that. I'm not saying there isn't a loss of depth, but it's not affecting the Bucks in the way we'd typically expect depth to affect a team with an MVP level player.

Re: health. There's certainly something to that. I expect the Bucks would look a lot better if they had their best version of Brook Lopez.

Re: 3 all-star names doesn't mean a better supporting cast. That's true, on the other hand those other two stars led the team to victory in the 2nd round of the playoffs without Giannis, and there's evidence this season as well that they can lead the offense pretty dang well without Giannis.

Re: Celtics role players delivery more this year especially on defense. Can't disagree with that, but I'd again emphasize that the Celtics +/- trends don't look like a team that's winning based on depth instead of stars. This isn't a situation where like, say, the Melo Nuggets where the team seemed to do okay without him until he left...and then they looked even better. For all the performance of the other players, Tatum's numbers stand out head and shoulders above everyone else - and frankly the same as been true for years now.

For perspective, here are the raw +/- totals of the Celtics over the past 4 years all together:

Tatum +1681
Smart +983
Brown +753

This is a just plain a huge lead. I don't know if you'll find any lead like from any other team among a core that's been together the whole time.

Re: just higher than you on +/-. Yup, and I should probably let you be.

I’m on mobile now, but could you check the Jazz for me? I could see Gobert with a similar lead