Top 25 peaks of the 2001-25: #3-#4 Spots

[One_and_Done]

Many, many players carried their teams more than 06 Kobe.

Based on what measure?

The 06 Lakers were a 45-win team, +2.2 on O, -0.5 on D. They started Smush Parker, Lamar Odom, Chris Mihm, and either Kwame Brown or Brian Cook alongside Kobe. They had Sasha Vujacic, Luke Walton and the largely-useless remnants of Devean George off the bench.

That was a pretty serious carry job.

I mean, Odom was a borderline all-star type player, so that's already pretty different from alot of carry jobs right there. If he'd stayed with the Heat with Wade instead of going West and eventually becoming a 6th man, then he'd have made multiple all-star appearances.

Mihm was a decent enough 5 before he got injured, and while Kwame gets alot of hate for being a failure of a #1 pick he too was a decent player. He was in the league for 12 years, and started almost half his games. He was repeatedly signed to decent money contracts by teams who valued him. No, he wasn't a good player, but he matured into a solid-ish one.

Smush wasn't good, but he's one guy. Cook was a bench player, but he basically played that role with his 19mpg. L.Walton was an ok bench player. Sasha only had 1 good year, and this wasn't it, but he was their 9th man.

I grant you it wasn't a good support cast, but calling it a historical carry job is way too far. Odom alone negates that claim. They also won only 45 games.

[ReggiesKnicks]

Something notable with Giannis. From 19-24 the Bucks were 281-118 with him, and only 40-35 without. That's the difference between a 57 and 43 win team, despite alot of RS coasting and the Doc Rivers disaster.

Doc Rivers coached literally half a season in the time frame you mention.

[ReggiesKnicks]

Many, many players carried their teams more than 06 Kobe.

Based on what measure?

The 06 Lakers were a 45-win team, +2.2 on O, -0.5 on D. They started Smush Parker, Lamar Odom, Chris Mihm, and either Kwame Brown or Brian Cook alongside Kobe. They had Sasha Vujacic, Luke Walton and the largely-useless remnants of Devean George off the bench.

That was a pretty serious carry job.

Also we have this data point.

Kobe + Odom: +5.3
Kobe, No Odom: +1.6
No Kobe, Odom: -7.7
No Kobe, No Odom: -11.4

All of these are on the smaller side for sample sizes, but the point remains. The Lakers had just 1 player with a positive BPM outside of Kobe in Odom.

Kwame Brown put up -2.5 BPM, yet is being argued as a decent player? That can't be right.

[One_and_Done]

Many, many players carried their teams more than 06 Kobe.

Based on what measure?

The 06 Lakers were a 45-win team, +2.2 on O, -0.5 on D. They started Smush Parker, Lamar Odom, Chris Mihm, and either Kwame Brown or Brian Cook alongside Kobe. They had Sasha Vujacic, Luke Walton and the largely-useless remnants of Devean George off the bench.

That was a pretty serious carry job.

Also we have this data point.

Kobe + Odom: +5.3
Kobe, No Odom: +1.6
No Kobe, Odom: -7.7
No Kobe, No Odom: -11.4

All of these are on the smaller side for sample sizes, but the point remains. The Lakers had just 1 player with a positive BPM outside of Kobe in Odom.

Kwame Brown put up -2.5 BPM, yet is being argued as a decent player? That can't be right.

If I cared about BPM that would be a compelling point to me.

[ReggiesKnicks]

Based on what measure?

The 06 Lakers were a 45-win team, +2.2 on O, -0.5 on D. They started Smush Parker, Lamar Odom, Chris Mihm, and either Kwame Brown or Brian Cook alongside Kobe. They had Sasha Vujacic, Luke Walton and the largely-useless remnants of Devean George off the bench.

That was a pretty serious carry job.

Also we have this data point.

Kobe + Odom: +5.3
Kobe, No Odom: +1.6
No Kobe, Odom: -7.7
No Kobe, No Odom: -11.4

All of these are on the smaller side for sample sizes, but the point remains. The Lakers had just 1 player with a positive BPM outside of Kobe in Odom.

Kwame Brown put up -2.5 BPM, yet is being argued as a decent player? That can't be right.

If I cared about BPM that would be a compelling point to me.

Unfortunetely you are the epitome of someone who is impossible to engage with in a meaningful way. Please add me to your foe list, I will do the same, and we can be two ships passing in the night.

[One_and_Done]

I've never felt the need to block anyone, but you're welcome to block me.

When we look at stats like TS% or pp100, they can be misleading. Like all stats they need context. However, they are still measuring something that has quantifiable certainty. For instance, if a guy shot 90% from the FT line, we know he could shoot free throws. If a guy shot 40% on seven 3s a game, we can feel confidence they were a decent shooter.

These 'one number' stats like BPM, whether you call them 'advanced' or not, are completely different. Having a good/bad BPM might mean something, or it might mean nothing.

To illustrate this, let’s take the example of a player who “hypothetically” carried a team of scrubs, let's call him Jebron. Just by virtue of the fact that 4 other guys are always sharing the court with a guy who is able to carry them and make the team look good, is going to warp the on-off numbers completely. Then imagine the guy is playing over 40 minutes a game, and the minutes where he’s off are often minutes where the coach had thrown the towel in. Add random variance and small sample size theatre, and the plus-minus numbers for that team won’t mean much. Similarly, you’re going to get more win-shares on a team that wins more.

I could keep going. Imagine the opposite of Jebron, a guy who actively hurts all the players he plays with and thus renders plus minus type numbers void, let's call him Dicky Ravis. Role, play style, quality of back-ups, etc. So many things can warp what these numbers are trying to measure to the point that they are just not reliable.

I don't see these numbers as a persuasive approach to rating players.

[lessthanjake]

Is there anything positive Austin Rivers can provide on the basketball court? He's a poor defender, weak shooter, bad ball-handler. He can't even make FTs on acceptable rate.

What makes him "fine" 6th man for the playoff team? Can you enlighten me?

I wouldn't be happy to have Austin Rivers as my 6th man (though we're talking RS now, not playoffs), but for one of the greatest peaks of all-time? It's not the flex you seem to think it is.

T-Mac led a Magic team to 42 wins, and took the Pistons to 7 games, and his 6 highest minute guys were Darrell Armstrong, Patrick Garrity, Mike Miller (49 games only), J.Vaughn, Fat Shawn Kemp, and Andrew DeClercq. Not one of those guys aside from Miller was a starting calibre guy, and the last 2 didn't even belong in the league. I'd flex that more than the 48 wins and 1-4 first round loss Jokic had.

There are tonnes of guys unnamed yet who have carried worse support casts to better results than Jokic did in 22. Giannis led the Bucks to 60 wins and his 6th and 7th highest minute guys were Snell and Connaughton (and they were even better in the 72 games Giannis did play, going only 4-6 without him).

I think this talk of Austin Rivers is missing the forest for the trees in terms of how awful the 2022 Nuggets were. Was Austin Rivers fit to be a 6th man on an NBA team that makes the playoffs? Absolutely not. But if the rest of the roster was strong (i.e. if it actually was the “top heavy team” that One_and_Done claimed it was), then Austin Rivers being a massive roster weakness might be something we’d think the rest of the team could potentially overcome. However, the rest of the team was not strong. Quite the opposite.

They didn’t just have Austin Rivers as their 6th man. The starting lineup around Jokic was Aaron Gordon, Will Barton, Monte Morris, and a 35-year-old Jeff Green. And the bench guys with the most minutes besides Austin Rivers were Facundo Campazzo, Bones Hyland, JaMychal Green, and Zeke Nnaji.

Aaron Gordon is an actually pretty good player, though he was definitely not as good at that point as he’s become since then. And beyond Gordon, the roster was an absolute disaster. Let’s start with the other starters and look at what they’ve done after that year. Will Barton would barely play NBA basketball after that year, washing out of the league after playing 56 more games. Monte Morris has not been able to find himself real rotation minutes, even as a backup PG. Jeff Green has hung around in the NBA as a low-minute bench player, continuing his run as a long-standing NBA player who consistently has amongst the worst RAPMs in NBA history. Then we get to Austin Rivers as an absolutely awful sixth man, who washed out of the league in a couple years. Beyond that, we have Bones Hyland, who basically cannot get NBA minutes at all after leaving Denver. Facundo Campazzo played a grand total of 52 minutes in the NBA after that year, having to leave for another league. JaMychal Green played one more year in the NBA as a bench player and then was left unsigned. And Zeke Nnaji is in the running for worst backup center in the league.

This was a team whose rotation was largely made up of guys who washed out of the league or ceased getting meaningful minutes soon thereafter. People like to exaggerate and talk about superstars playing with G-Leaguers, but the 2022 Nuggets are actually an example where that description is pretty accurate. The rotation was genuinely Jokic, Gordon (not as good as he later became), and a bunch of guys who don’t even belong in the NBA. It was an utterly awful supporting cast. Just at a completely different level than the bad supporting casts we talk about with other great players. Teams that are tanking only wish they could have a rotation this bad (minus Jokic of course). Taking that team to 48 wins was an absolutely ridiculous carry. And any attempt to act like that team was not absolutely pathetic is just completely crazy. It was a different level of bad, well beyond what we typically talk about when we label a star’s supporting cast as bad (including the examples given in the above-quoted post).

[lessthanjake]

I've never felt the need to block anyone, but you're welcome to block me.

When we look at stats like TS% or pp100, they can be misleading. Like all stats they need context. However, they are still measuring something that has quantifiable certainty. For instance, if a guy shot 90% from the FT line, we know he could shoot free throws. If a guy shot 40% on seven 3s a game, we can feel confidence they were a decent shooter.

These 'one number' stats like BPM, whether you call them 'advanced' or not, are completely different. Having a good/bad BPM might mean something, or it might mean nothing.

To illustrate this, let’s take the example of a player who “hypothetically” carried a team of scrubs, let's call him Jebron. Just by virtue of the fact that 4 other guys are always sharing the court with a guy who is able to carry them and make the team look good, is going to warp the on-off numbers completely. Then imagine the guy is playing over 40 minutes a game, and the minutes where he’s off are often minutes where the coach had thrown the towel in. Add random variance and small sample size theatre, and the plus-minus numbers for that team won’t mean much. Similarly, you’re going to get more win-shares on a team that wins more.

I could keep going. Imagine the opposite of Jebron, a guy who actively hurts all the players he plays with and thus renders plus minus type numbers void, let's call him Dicky Ravis. Role, play style, quality of back-ups, etc. So many things can warp what these numbers are trying to measure to the point that they are just not reliable.

I don't see these numbers as a persuasive approach to rating players.

You know that BPM is not a stat that’s based on plus-minus or on-off, right? You say BPM “might mean nothing,” and follow that up by saying “[t]o illustrate this” and then give an example focused on how context “is going to warp the on-off numbers completely.” It seems like you don’t know what BPM is and just saw “Plus Minus” in the name and figured it was an impact stat that uses on and off numbers. It is a purely box metric.

[70sFan]

What are we even discussing here? Austin Rivers is called a better player than Pat Connaughton. Old Jeff Green is more valued than old Horace Grant and Ron Harper. Will Barton is called a good starter. BPM is trashed without knowing what it is (and I actually don't value boxscore estimates much, but I actually know what they are). Kwame Brown, one of the worst starters in the league for a decade, is somehow a reason the team should win more than 45 games.

All of that of course isn't backed up by anything. All of that isn't even vibe-based discussion, it's all dictated to presuppositions. I am still waiting for explaination what makes Austin Rivers a solid 6th man, the guy should never play in the league with all respect to him.

[ReggiesKnicks]

4. 2021 Giannis Antetokounmpo
This is where it's starting to get really hard. I had a hard time deciding between him and Shaq and the 2000 version of Shaq would have been ahead of Giannis for sure, but I can't quite get there for 2001. Shaq just never really played elite defense except for the 99/00 season. Meanwhile, Giannis here is at the absolute peak of his powers on both ends and puts on a cape and goes superhuman mode when the Finals come around. It's true that the Lakers had an all-time dominant playoff run in 2001, but that's more because it's the one year that Kobe played up to Shaq's level than it is that Shaq did anything particularly special.

This is the hard part, I agree. I haven't been blown away by Giannis like yourself or others, and perhaps it is the division rivalry and thinking Giannis is a tad-bit overrated offensively, but I think I have been unfair.

Giannis stacks up incredibly well in +/- when looking at his minutes with Jrue, better than Kobe/Shaq +/- together and comparable to Jokic/Murray and Garnett/Cassell.

4. Kobe Bryant 2006
I know most people here won't see him as high as I do. But Kobe that year was special. The West was talented for sure, and Kobe drove that team to the playoffs and took the Suns to 7 games. One of his best years ts% wise, and given his offensive explosions I gotta say I was super impressed that year with his production.

Also his lower TOV% means per possession he was actually more efficient than even his ts% suggests.

On defense he took possessions off - that's only normal with such offensive burden. However, when needed he still displayed his man to man agressive and great defense.

I don't think many players could replicate his impact on winning so many games with a very bad roster. Kobe was special that year.

Maybe Giannis 21 and KG 04 should go in here but I just think Kobe did very well given his conditions and I don't see many guys replicating him as an offensive force.

Kobe's 2006 season was incredible as an offensive carry-job, but even focusing on this and disregarding the fact that the approach Kobe had to take in 2006 didn't provide championship equity, other players you mentioned (and I will reference) show more impact.

Here are Kobe's On-Court and On/Off +/-, along with BPM in 2006: +4.6 / +12.5 / 8.0
The impressive thing here, to contextualize the numbers, is Kobe had one player, Lamar Odom, with positive BPM. Use any metric you want to measure teammate strength, but this fact sheds light into the lack of top-end talent (Think Top 100-150 players in the NBA) the 2006 Lakers had to work with. Teams with only 2 Top 100-150 players don't usually make a lot of noise.

Compare this to Kevin Garnett. Here are his annual averages from 2000-2003, 4 seasons, with Kobe's Below (+/-, On/Off, VORP)

+5.2 / +13.6 / 7.3
+4.6 / +12.5 / 8.0

Garnett, over a 4-year span, is similar (yet perhaps slightly more impressive) than Kobe Bryant in terms of "carry jobs". Imagine that, a 4-year span matching Kobe's 1-year span. This doesn't even include Garnett's peak, 2004.

Now, it is true Garnett may have had more talent, but it isn't a meaningful Gap.

2000: Szczerbiak squeaks in at 0.9 BPM and the Timberwolves win 50 games
2001: Szczerbiak squeaks in at 0.4 BPM and the Timberwolves win 47 games
2002: Szczerbiak gets in at 1.4 BPM and the Timberwolves win 50 games
2003: Szczerbiak is the ONLY teammate with a positive BPM (Far less than Odom and only plays 52 games)

And that is where we should focus, 2003. Garnett's best teammate was 52-games of Wally Szczerbiak, no other player with a positive BPM. The Timberwolves win 51-games. The Timberwolves are the best clutch team in the NBA. They go 27-14, leading the league in clutch Win% and Total Wins, 2nd in +/-. That's it.

Kobe Bryant and the Lakers were instead 18-23 in clutch games, boasting a -13 +/- in those games (Minnesota had +52).

Now, you may be thinking variance, noise, sample size. Why yes, I have your answer below, which includes the Timberwolves from 2000-2003 Clutch Win% and Total +/- ranking

2000: 29-18 (61.7%), +107 (1st in +/-)
2001: 23-20 (53.5%), -18 (19th in +/-)
2002: 22-19 (53.7%), -17 (18th in +/-)
2003: 27-14, +52 (2nd in +/-)

All of these seasons eclipse the 2006 Lakers in Win%. Coincidence? You can be the judge there.

And what about Russell Westbrook in 2017? That carry job was incredible. He had one teammate with a positive BPM, Enes Freedom, who played just 1500 minutes.

2006 Lakers
45-37

2017 Thunder
47-35

Below was Kobe's +/-, On/Off and VORP

+4.6 / +12.5 / 8.0

Here is Westbrooks:
+4.0 / +12.5 / 9.3

And here, the statistical comparison is easier since there isn't a gulf of defensive impact gap between Kobe and Westbrook like there was with Garnett and either.

Westbrook
USG%: 41.7
VORP: 9.3
AST%: 57.3
Per-100: 45-15-15-2
Assist-Turnover Ratio: 1.9

Kobe
USG%: 38.7
VORP: 8.0
AST%: 24.1
Per-100: 46-7-6-2
Assist-Turnover Ratio: 1.5

Kobe's turnover economy is valuable on good teams, but referencing his turnover economy on teams lacking playmakers isn't a feather in his cap. If you are truly going to argue for 2006 Kobe, his turnover economy and inability to take on a large burden of offensive playmaking, specifically passing, is actually a weakness.

I am going to assume I have Kobe as a better player than Westbrook, though I haven't fully fleshed out the comparison. However, I wouldn't try to argue 2006 Kobe as being better than 2017 Westbrook as the reason why Kobe > Westbrook, which is another reason having 2006 Kobe this high is a travesty.

4. Dwyane Wade (2006)

I was writing up an entry here for 2004 Garnett, and ended up changing my vote mid-post, because I started talking about why I put Garnett above 2006 Wade and I realized I didn’t really buy my own reasoning.

The crux of Wade’s case here is that he had an incredible playoff run, leading the Heat to the title. In my view, that Heat team had little business winning the NBA title. Shaq was still a good player, but he was pretty diminished by the end. Meanwhile, the Heat gave their second-most playoff minutes (and almost as many shots as Shaq) to Antoine Walker—who IMO is arguably the worst consistent starter in NBA history, with just awful impact data, combined with my eye test thinking he was just terrible even back then. Jason Williams, Udonis Haslem, and James Posey ranged from neutral to somewhat negative players IMO. They did benefit from depth, in that their 7th and 8th men were old Gary Payton and old Alonzo Mourning, who were both positive players in the limited minutes they got. Overall, in my view, that team was not really a championship-quality team. And yet Wade led them to a title. I found it shocking at the time, and I still do.

This wasn’t an easy road either. The first two rounds were not against overly difficult opponents, but I wouldn’t say the Kidd/Carter/Jefferson Nets were minnows either. More importantly, the Heat faced the Pistons in the conference finals. This was a team that had won the title in 2004, and had lost a close Game 7 in the Finals in 2005. They had also just had their best regular season, winning 64 games. This was the title favorite. And Wade absolutely torched them, putting up a 68.4% TS% for the series. The Heat then played the 60-win, 6 SRS Mavericks, who had just gotten done beating the Spurs and Suns and looked pretty destined to win the title. And Wade torched them too, carrying the Heat with 35 points a game, including just dominating the last four games that the Heat won after going down 0-2. Basically, Wade dominated two great teams that I don’t think the Heat had any business beating. FWIW, he also had a +22.2 on-off in the playoffs, though I don’t value that much due to the tiny sample size.

So I think 2006 Wade had a playoffs that is on my short list for most impressive title runs by a player. But what about the regular season? Well, it was still really good. He led the league in RAPTOR, and was basically in a tight group of a few players near the top of the league in other metrics. He had a fantastic +15.2 on-off. He definitely wasn’t clearly the best player in the regular season. But he was in the conversation. And when combined with one of the most impressive playoff runs ever, I find it very compelling.

As I mentioned, I was going to put 2004 Garnett here. I certainly think it’d be relatively straightforward to conclude that 2004 Garnett generally had more impact per 100 possessions than 2006 Wade. But Wade was still very impactful, and 2004 Garnett simply did not have the playoff run that 2006 Wade had. So, overall, I just find it difficult to conclude that 2004 Garnett was actually “greater” than 2006 Wade.

The actual question for me is whether 2006 Wade should be even higher. Is 2001 Shaq better than this? I guess my logic here is that Shaq had a slightly better argument for being the league’s best player in the regular season than Wade did, and they both were extremely impressive in the playoffs (with Shaq leading his team to a 15-1 playoff record, while Wade led a team to a title that had no business winning it). That shorthand logic ends up with Shaq a bit ahead. But I struggle with the conclusion that 2001 Shaq’s regular season actually was better than 2006 Wade’s, since the SRS of the two teams was virtually identical and I think Shaq had the better supporting cast. That said, I do look at surrounding years and see Shaq looking better than Wade, and that gives me some info about their individual level in these particular years. So, while I’m not certain 2006 Wade doesn’t deserve to be #3 here, I am not quite ready to do it.

Wade vs Garnett

I fail to see how Wade was clearly better in the 2006 Post-Season than Garnett. But, before we get to this, the gap in the regular season is incredibly large.

2004 Garnett
Total +/- : 614 (1st in NBA)
Garnett+Cassell (RS 2004+2005): +9.1
Garnett+Cassell (RS 2004 Only): +11.3
Garnett+Cassell (PS 2004): +6.4
10.2 BPM (League Leader)
18.3 WS (League Leader)
30.0 DRB% (League Leader)
1.9 Assist:Turnover Ratio
2.0 STL%
4.0 BLK%

33-19-7-2-3 Per-100 Statline on 113 TS+

2006 Wade
Total +/-: 485 (10th in NBA)
Wade+Shaq (RS 2005+2006): +8.9
Wade+Shaq (RS 2006 Only): +10.3
Wade+Shaq (PS 2006): +6.0
7.7 BPM
7.1 VORP
33 AST%
1.8 Assist:Turnover Ratio
2.6 STL%
1.5 BLK%

37-8-9-3-1 Per-100 Statline on 148 TS+

There is clearly a monumental gap between the two players in the Regular Season. That is for certain.

However, when we get to the post-season, Garnett was just as good as he was in the regular season, which was a dominant, ATG MVP season. Both players produced a similar +/- in the post-season when sharing the court with their robin, without acknowledging the strength of their robins or opponents.

Furthermore, you mention On-Off for Wade at +15.2 in the Regular Season. Garnett's dwarfs this at +20.7. Wade has an incredible On/Off (Small sample) in the post-season at +22.2, yet Garnett again bests this at +26.7.

I see nothing in your post to suggest Wade was better than Garnett. Garnett was significantly better in the regular season and, at worst, similar in the post-season.

[lessthanjake]

4. Dwyane Wade (2006)

I was writing up an entry here for 2004 Garnett, and ended up changing my vote mid-post, because I started talking about why I put Garnett above 2006 Wade and I realized I didn’t really buy my own reasoning.

The crux of Wade’s case here is that he had an incredible playoff run, leading the Heat to the title. In my view, that Heat team had little business winning the NBA title. Shaq was still a good player, but he was pretty diminished by the end. Meanwhile, the Heat gave their second-most playoff minutes (and almost as many shots as Shaq) to Antoine Walker—who IMO is arguably the worst consistent starter in NBA history, with just awful impact data, combined with my eye test thinking he was just terrible even back then. Jason Williams, Udonis Haslem, and James Posey ranged from neutral to somewhat negative players IMO. They did benefit from depth, in that their 7th and 8th men were old Gary Payton and old Alonzo Mourning, who were both positive players in the limited minutes they got. Overall, in my view, that team was not really a championship-quality team. And yet Wade led them to a title. I found it shocking at the time, and I still do.

This wasn’t an easy road either. The first two rounds were not against overly difficult opponents, but I wouldn’t say the Kidd/Carter/Jefferson Nets were minnows either. More importantly, the Heat faced the Pistons in the conference finals. This was a team that had won the title in 2004, and had lost a close Game 7 in the Finals in 2005. They had also just had their best regular season, winning 64 games. This was the title favorite. And Wade absolutely torched them, putting up a 68.4% TS% for the series. The Heat then played the 60-win, 6 SRS Mavericks, who had just gotten done beating the Spurs and Suns and looked pretty destined to win the title. And Wade torched them too, carrying the Heat with 35 points a game, including just dominating the last four games that the Heat won after going down 0-2. Basically, Wade dominated two great teams that I don’t think the Heat had any business beating. FWIW, he also had a +22.2 on-off in the playoffs, though I don’t value that much due to the tiny sample size.

So I think 2006 Wade had a playoffs that is on my short list for most impressive title runs by a player. But what about the regular season? Well, it was still really good. He led the league in RAPTOR, and was basically in a tight group of a few players near the top of the league in other metrics. He had a fantastic +15.2 on-off. He definitely wasn’t clearly the best player in the regular season. But he was in the conversation. And when combined with one of the most impressive playoff runs ever, I find it very compelling.

As I mentioned, I was going to put 2004 Garnett here. I certainly think it’d be relatively straightforward to conclude that 2004 Garnett generally had more impact per 100 possessions than 2006 Wade. But Wade was still very impactful, and 2004 Garnett simply did not have the playoff run that 2006 Wade had. So, overall, I just find it difficult to conclude that 2004 Garnett was actually “greater” than 2006 Wade.

The actual question for me is whether 2006 Wade should be even higher. Is 2001 Shaq better than this? I guess my logic here is that Shaq had a slightly better argument for being the league’s best player in the regular season than Wade did, and they both were extremely impressive in the playoffs (with Shaq leading his team to a 15-1 playoff record, while Wade led a team to a title that had no business winning it). That shorthand logic ends up with Shaq a bit ahead. But I struggle with the conclusion that 2001 Shaq’s regular season actually was better than 2006 Wade’s, since the SRS of the two teams was virtually identical and I think Shaq had the better supporting cast. That said, I do look at surrounding years and see Shaq looking better than Wade, and that gives me some info about their individual level in these particular years. So, while I’m not certain 2006 Wade doesn’t deserve to be #3 here, I am not quite ready to do it.

Wade vs Garnett

I fail to see how Wade was clearly better in the 2006 Post-Season than Garnett. But, before we get to this, the gap in the regular season is incredibly large.

2004 Garnett
Total +/- : 614 (1st in NBA)
Garnett+Cassell (RS 2004+2005): +9.1
Garnett+Cassell (RS 2004 Only): +11.3
Garnett+Cassell (PS 2004): +6.4
10.2 BPM (League Leader)
18.3 WS (League Leader)
30.0 DRB% (League Leader)
1.9 Assist:Turnover Ratio
2.0 STL%
4.0 BLK%

33-19-7-2-3 Per-100 Statline on 113 TS+

2006 Wade
Total +/-: 485 (10th in NBA)
Wade+Shaq (RS 2005+2006): +8.9
Wade+Shaq (RS 2006 Only): +10.3
Wade+Shaq (PS 2006): +6.0
7.7 BPM
7.1 VORP
33 AST%
1.8 Assist:Turnover Ratio
2.6 STL%
1.5 BLK%

37-8-9-3-1 Per-100 Statline on 148 TS+

There is clearly a monumental gap between the two players in the Regular Season. That is for certain.

However, when we get to the post-season, Garnett was just as good as he was in the regular season, which was a dominant, ATG MVP season. Both players produced a similar +/- in the post-season when sharing the court with their robin, without acknowledging the strength of their robins or opponents.

Furthermore, you mention On-Off for Wade at +15.2 in the Regular Season. Garnett's dwarfs this at +20.7. Wade has an incredible On/Off (Small sample) in the post-season at +22.2, yet Garnett again bests this at +26.7.

I see nothing in your post to suggest Wade was better than Garnett. Garnett was significantly better in the regular season and, at worst, similar in the post-season.

At the outset, I’ll say that a big reason I think Wade was substantially greater in the 2006 playoffs than Garnett was in the 2004 playoffs is that Wade dominated multiple great opponents to drag his team to the title, while Garnett led his team to a six-game conference finals loss (to the team that easily lost in the finals). That weighs heavily for purposes of assessing the “greatness” of their years, even if you might think that Garnett generally had more impact per 100 possessions than Wade did.

And even leaving aside the massive gap in what they led their teams to, at an individual level I think Wade was definitely better in the playoffs, and I certainly don’t think any series from Garnett in 2004 was as impressive as Wade’s series against the Pistons and Mavericks. Garnett was definitely a beast defensively, but he struggled with his scoring in those playoffs, putting up a negative rTS%.

You assert that Garnett was the better player in the playoffs, but without much reasoning for that beyond some reference to raw plus-minus data and a naked assertion that he “was just as good as he was in the regular season.” For reference, Garnett had a +2.9 EPM in the 2004 playoffs. That was not only a large drop off from the +6.8 he put up in the regular season, but was also miles below Wade’s +7.3 in the 2006 playoffs. Similarly, Garnett put up a 6.5 BPM in the 2004 playoffs, which was again a large drop off from the 10.2 he had in the regular season and far below the 9.3 that Wade had in the 2006 playoffs. Similarly, Garnett put up 0.163 WS/48 in the 2004 playoffs, way below his 0.272 in the regular season and Wade’s 0.240 in the 2006 playoffs.

So yeah, I think it’s pretty straightforward to conclude that Garnett was not as good in the 2004 playoffs as he was in the 2004 regular season, and that Wade was significantly better in the 2006 playoffs than Garnett was in the 2004 playoffs. The question is just whether Garnett being the more impactful regular season player overcomes that, for purposes of the “greatness” of the year. As I mentioned in my voting post, I initially started writing my voting post intending to argue exactly that, but then I realized while writing it that I didn’t buy it. Wade’s substantially better playoffs and the fact that he dragged a team to a title that I think had no business winning it simply weighs more highly to me.

[Doctor MJ]

So yeah, I think it’s pretty straightforward to conclude that Garnett was not as good in the 2004 playoffs as he was in the 2004 regular season, and that Wade was significantly better in the 2006 playoffs than Garnett was in the 2004 playoffs. The question is just whether Garnett being the more impactful regular season player overcomes that, for purposes of the “greatness” of the year. As I mentioned in my voting post, I initially started writing my voting post intending to argue exactly that, but then I realized while writing it that I didn’t buy it. Wade’s substantially better playoffs and the fact that he dragged a team to a title that I think had no business winning it simply weighs more highly to me.

So you & Reggie are having a good conversation and I apologize for jumping in and just responding to this.

When you say "pretty straightforward to conclude that Garnett was not as good in the 2004 playoffs as he was in the 2004 regular season" - forgetting the specific context of the debate with Wade - it makes me nervous.

If it's straight forward to conclude Garnett was worse in the playoffs, we should be able to explain the why & how by means more specific than all-in-ones, right?

If we just look at the traditional box score for KG in '03-04:

RS: 24.2 PPG, 13.9 RPG, 5.0 APG 1.5 SPG 2.2 BPG
PS: 24.3 PPG, 14.6 RPG, 5.1 APG 1.3 SPG 2.3 BPG

Just looking at that, it certainly isn't straight forward that he was worse.

And going by the simple +/-:

RS: On +9.8, On-Off +20.7
PS: On +2.5, On-Off +26.7

So, less separation from the other team when he's on that in the RS, but the small sample size of the On-Off only favors him more.

So what's left? Pretty much just efficiency, right? His TS% drops from 54.7 to 51.3, his TOV% rises from 10.5 to 14.9.

Now, efficiency is important don't get me wrong, but I do think we need to be careful when we talk about a player "getting worse in the playoffs" based on situations where his team actually relied upon him even more in the playoffs. Remember, it's not just that Garnett played even more with higher playmaking primacy in the playoffs against tougher competition, it's that Cassell gets hurt and drops off, and does so specifically when they go up against their toughest competition - he played 64 minutes against the Lakers while Garnett played 264.

And how's Garnett's efficiency without any all-star level help (because let's be real, Spree & Wally weren't that) in that series compared to Kobe playing with Shaq, Malone, Payton & co? Basically the same (Kobe .519, Garnett .518).

To be clear, I say none of this looking to assert that Garnett was the level of scorer of Kobe or Wade, or even as good of an offensive player, but if we're talking about a guy being the primary scorer & facilitator on a team with much less talent, and he's still score about as efficiently as his opponent for whom scoring is his THING, how exactly are we thinking Garnett disappointed compared to the regular season?

Finally as I say all of this, I do think it's fine to say that the Timberwolves offensive scheme was problematic and destined to not scale as well as other schemes. We should remember, for example, that this was a team who was 27th in 3PA while playing in a league where even being #1 in 3PA was literally a sub-optimal scheme by a good margin knowing what we know now.

This then to say: It's entirely possible that if you just gave Minny a coach with Pop's level of awareness (let alone the awareness levels of a competent 2025 coach) and turned their long twos into 3's, we're talking about a team that wins the chip, and if that had happened, would any of us really see Garnett as being worse in the playoffs? Unlikely I think.

[lessthanjake]

So yeah, I think it’s pretty straightforward to conclude that Garnett was not as good in the 2004 playoffs as he was in the 2004 regular season, and that Wade was significantly better in the 2006 playoffs than Garnett was in the 2004 playoffs. The question is just whether Garnett being the more impactful regular season player overcomes that, for purposes of the “greatness” of the year. As I mentioned in my voting post, I initially started writing my voting post intending to argue exactly that, but then I realized while writing it that I didn’t buy it. Wade’s substantially better playoffs and the fact that he dragged a team to a title that I think had no business winning it simply weighs more highly to me.

So you & Reggie are having a good conversation and I apologize for jumping in and just responding to this.

When you say "pretty straightforward to conclude that Garnett was not as good in the 2004 playoffs as he was in the 2004 regular season" - forgetting the specific context of the debate with Wade - it makes me nervous.

If it's straight forward to conclude Garnett was worse in the playoffs, we should be able to explain the why & how by means more specific than all-in-ones, right?

If we just look at the traditional box score for KG in '03-04:

RS: 24.2 PPG, 13.9 RPG, 5.0 APG 1.5 SPG 2.2 BPG
PS: 24.3 PPG, 14.6 RPG, 5.1 APG 1.3 SPG 2.3 BPG

Just looking at that, it certainly isn't straight forward that he was worse.

And going by the simple +/-:

RS: On +9.8, On-Off +20.7
PS: On +2.5, On-Off +26.7

So, less separation from the other team when he's on that in the RS, but the small sample size of the On-Off only favors him more.

So what's left? Pretty much just efficiency, right? His TS% drops from 54.7 to 51.3, his TOV% rises from 10.5 to 14.9.

Now, efficiency is important don't get me wrong, but I do think we need to be careful when we talk about a player "getting worse in the playoffs" based on situations where his team actually relied upon him even more in the playoffs. Remember, it's not just that Garnett played even more with higher playmaking primacy in the playoffs against tougher competition, it's that Cassell gets hurt and drops off, and does so specifically when they go up against their toughest competition - he played 64 minutes against the Lakers while Garnett played 264.

And how's Garnett's efficiency without any all-star level help (because let's be real, Spree & Wally weren't that) in that series compared to Kobe playing with Shaq, Malone, Payton & co? Basically the same (Kobe .519, Garnett .518).

To be clear, I say none of this looking to assert that Garnett was the level of scorer of Kobe or Wade, or even as good of an offensive player, but if we're talking about a guy being the primary scorer & facilitator on a team with much less talent, and he's still score about as efficiently as his opponent for whom scoring is his THING, how exactly are we thinking Garnett disappointed compared to the regular season?

Finally as I say all of this, I do think it's fine to say that the Timberwolves offensive scheme was problematic and destined to not scale as well as other schemes. We should remember, for example, that this was a team who was 27th in 3PA while playing in a league where even being #1 in 3PA was literally a sub-optimal scheme by a good margin knowing what we know now.

This then to say: It's entirely possible that if you just gave Minny a coach with Pop's level of awareness (let alone the awareness levels of a competent 2025 coach) and turned their long twos into 3's, we're talking about a team that wins the chip, and if that had happened, would any of us really see Garnett as being worse in the playoffs? Unlikely I think.

These are not unfair points, but just a few things I’d note in response:

1. The difference in efficiency is pretty significant IMO. He went down to a negative rTS%, and the increase in turnovers was pretty significant. This sort of thing is definitely enough to move a player down a tier in terms of how well they played.

2. You use per-game averages, but Garnett played over 4 more MPG in the playoffs than in the regular season. So, for instance, while the PPG in regular season and playoffs was basically the same, his points per 100 possessions went from 33.2 to 29.9. So his per-possession output went down and his efficiency also went down. And now I think we’re definitely starting to see something that moves a player down significantly (and we also are likely seeing a lot of what ended up causing the downturn in the box and impact data I cited). Granted, he should get credit for the fact that he played big minutes in the playoffs. That does mitigate some of his downturn in per-possession quality from regular season to playoffs. But Wade played a lot of playoff minutes too, so this isn’t much of a mitigating factor in a comparison between 2004 Garnett and 2006 Wade in the playoffs.

3. You make a comparison with Kobe’s efficiency in the Lakers series. And I understand why you did. But Kobe did not have a good 2004 playoffs. I definitely don’t think “scored with 2004 playoff Kobe efficiency” is a positive factor in a comparison with 2006 Wade. And, even leaving that aside, Kobe was facing a slightly better defense and one that gave up a notably lower regular season TS% (of course, that is in significant part a reflection of Garnett’s great defense, so credit to him for that). Kobe actually had a +3% rTS% in that series, while Garnett had an essentially exactly 0 rTS%. Having a notably lower rTS% than Kobe in a series that was in one of his least efficient playoffs is not great IMO.

4. I think you’re probably right that the Timberwolves could’ve potentially done better if they’d done things differently schematically. Of course, we could probably say that about every team in the NBA back then. And, in any event, for rankings like this, I tend to at least try not to focus much on speculation and hypotheticals, in favor of what actually happened. To some degree, speculation/hypotheticals are unavoidable, so I am not categorically opposed to them, but if my conclusion is that 2006 Wade had a greater season than 2004 Garnett but that 2004 Garnett was theoretically capable of having a greater season than 2006 Wade was capable of, then I will still put 2006 Wade ahead.

5. Just to briefly go down the speculation/hypothetical road, I don’t think I agree that we’d necessarily see 2004 Garnett as being just as good in the playoffs as 2006 Wade if the Timberwolves had won the title. I think it really depends on how well he actually played. I have 2006 Wade as being better in the playoffs than most superstars in their title runs. He was magnificent, and most superstars who win a title aren’t quite that good IMO. Garnett would’ve had to be genuinely very special in the playoffs for me to think he was as good as 2006 Wade. Maybe he’d have to have been that good for the Timberwolves to win a title, but I’m not sure that’s actually true if we are giving them a significant schematic upgrade.

[ReggiesKnicks]

4. Dwyane Wade (2006)

I was writing up an entry here for 2004 Garnett, and ended up changing my vote mid-post, because I started talking about why I put Garnett above 2006 Wade and I realized I didn’t really buy my own reasoning.

The crux of Wade’s case here is that he had an incredible playoff run, leading the Heat to the title. In my view, that Heat team had little business winning the NBA title. Shaq was still a good player, but he was pretty diminished by the end. Meanwhile, the Heat gave their second-most playoff minutes (and almost as many shots as Shaq) to Antoine Walker—who IMO is arguably the worst consistent starter in NBA history, with just awful impact data, combined with my eye test thinking he was just terrible even back then. Jason Williams, Udonis Haslem, and James Posey ranged from neutral to somewhat negative players IMO. They did benefit from depth, in that their 7th and 8th men were old Gary Payton and old Alonzo Mourning, who were both positive players in the limited minutes they got. Overall, in my view, that team was not really a championship-quality team. And yet Wade led them to a title. I found it shocking at the time, and I still do.

This wasn’t an easy road either. The first two rounds were not against overly difficult opponents, but I wouldn’t say the Kidd/Carter/Jefferson Nets were minnows either. More importantly, the Heat faced the Pistons in the conference finals. This was a team that had won the title in 2004, and had lost a close Game 7 in the Finals in 2005. They had also just had their best regular season, winning 64 games. This was the title favorite. And Wade absolutely torched them, putting up a 68.4% TS% for the series. The Heat then played the 60-win, 6 SRS Mavericks, who had just gotten done beating the Spurs and Suns and looked pretty destined to win the title. And Wade torched them too, carrying the Heat with 35 points a game, including just dominating the last four games that the Heat won after going down 0-2. Basically, Wade dominated two great teams that I don’t think the Heat had any business beating. FWIW, he also had a +22.2 on-off in the playoffs, though I don’t value that much due to the tiny sample size.

So I think 2006 Wade had a playoffs that is on my short list for most impressive title runs by a player. But what about the regular season? Well, it was still really good. He led the league in RAPTOR, and was basically in a tight group of a few players near the top of the league in other metrics. He had a fantastic +15.2 on-off. He definitely wasn’t clearly the best player in the regular season. But he was in the conversation. And when combined with one of the most impressive playoff runs ever, I find it very compelling.

As I mentioned, I was going to put 2004 Garnett here. I certainly think it’d be relatively straightforward to conclude that 2004 Garnett generally had more impact per 100 possessions than 2006 Wade. But Wade was still very impactful, and 2004 Garnett simply did not have the playoff run that 2006 Wade had. So, overall, I just find it difficult to conclude that 2004 Garnett was actually “greater” than 2006 Wade.

The actual question for me is whether 2006 Wade should be even higher. Is 2001 Shaq better than this? I guess my logic here is that Shaq had a slightly better argument for being the league’s best player in the regular season than Wade did, and they both were extremely impressive in the playoffs (with Shaq leading his team to a 15-1 playoff record, while Wade led a team to a title that had no business winning it). That shorthand logic ends up with Shaq a bit ahead. But I struggle with the conclusion that 2001 Shaq’s regular season actually was better than 2006 Wade’s, since the SRS of the two teams was virtually identical and I think Shaq had the better supporting cast. That said, I do look at surrounding years and see Shaq looking better than Wade, and that gives me some info about their individual level in these particular years. So, while I’m not certain 2006 Wade doesn’t deserve to be #3 here, I am not quite ready to do it.

Wade vs Garnett

I fail to see how Wade was clearly better in the 2006 Post-Season than Garnett. But, before we get to this, the gap in the regular season is incredibly large.

2004 Garnett
Total +/- : 614 (1st in NBA)
Garnett+Cassell (RS 2004+2005): +9.1
Garnett+Cassell (RS 2004 Only): +11.3
Garnett+Cassell (PS 2004): +6.4
10.2 BPM (League Leader)
18.3 WS (League Leader)
30.0 DRB% (League Leader)
1.9 Assist:Turnover Ratio
2.0 STL%
4.0 BLK%

33-19-7-2-3 Per-100 Statline on 113 TS+

2006 Wade
Total +/-: 485 (10th in NBA)
Wade+Shaq (RS 2005+2006): +8.9
Wade+Shaq (RS 2006 Only): +10.3
Wade+Shaq (PS 2006): +6.0
7.7 BPM
7.1 VORP
33 AST%
1.8 Assist:Turnover Ratio
2.6 STL%
1.5 BLK%

37-8-9-3-1 Per-100 Statline on 148 TS+

There is clearly a monumental gap between the two players in the Regular Season. That is for certain.

However, when we get to the post-season, Garnett was just as good as he was in the regular season, which was a dominant, ATG MVP season. Both players produced a similar +/- in the post-season when sharing the court with their robin, without acknowledging the strength of their robins or opponents.

Furthermore, you mention On-Off for Wade at +15.2 in the Regular Season. Garnett's dwarfs this at +20.7. Wade has an incredible On/Off (Small sample) in the post-season at +22.2, yet Garnett again bests this at +26.7.

I see nothing in your post to suggest Wade was better than Garnett. Garnett was significantly better in the regular season and, at worst, similar in the post-season.

At the outset, I’ll say that a big reason I think Wade was substantially greater in the 2006 playoffs than Garnett was in the 2004 playoffs is that Wade dominated multiple great opponents to drag his team to the title, while Garnett led his team to a six-game conference finals loss (to the team that easily lost in the finals). That weighs heavily for purposes of assessing the “greatness” of their years, even if you might think that Garnett generally had more impact per 100 possessions than Wade did.

Results-oriented thinking is valuable in the sense that results are important, but it can take away from actual player comparison and analysis.



Garnett 2004 Play-Offs On:

And even leaving aside the massive gap in what they led their teams to, at an individual level I think Wade was definitely better in the playoffs, and I certainly don’t think any series from Garnett in 2004 was as impressive as Wade’s series against the Pistons and Mavericks. Garnett was definitely a beast defensively, but he struggled with his scoring in those playoffs, putting up a negative rTS%.

Referencing rTS% for a player like Kevin Garnett as to why he struggled is showing a clear misunderstanding of a player like Kevin Garnett's ilk.

If we want to do a series-by-series analysis and analysis beyond struggled with his scoring in those playoffs as the leading measurement for Garnett's offense, we will see a clearer picture. Attempting to analyze offense through one lens, rTS%, isn't utilizing the important signals and tools we have at our disposal to properly assess these two all-time greats.

You assert that Garnett was the better player in the playoffs, but without much reasoning for that beyond some reference to raw plus-minus data and a naked assertion that he “was just as good as he was in the regular season.” For reference, Garnett had a +2.9 EPM in the 2004 playoffs. That was not only a large drop off from the +6.8 he put up in the regular season, but was also miles below Wade’s +7.3 in the 2006 playoffs. Similarly, Garnett put up a 6.5 BPM in the 2004 playoffs, which was again a large drop off from the 10.2 he had in the regular season and far below the 9.3 that Wade had in the 2006 playoffs. Similarly, Garnett put up 0.163 WS/48 in the 2004 playoffs, way below his 0.272 in the regular season and Wade’s 0.240 in the 2006 playoffs.

So yeah, I think it’s pretty straightforward to conclude that Garnett was not as good in the 2004 playoffs as he was in the 2004 regular season, and that Wade was significantly better in the 2006 playoffs than Garnett was in the 2004 playoffs. The question is just whether Garnett being the more impactful regular season player overcomes that, for purposes of the “greatness” of the year. As I mentioned in my voting post, I initially started writing my voting post intending to argue exactly that, but then I realized while writing it that I didn’t buy it. Wade’s substantially better playoffs and the fact that he dragged a team to a title that I think had no business winning it simply weighs more highly to me.

I think it is important to remember that there is a lot of noise in all-in-ones for small sample sizes. What I attempt to do with small sample sizes is make them larger, which increases their accuracy by increasing the sample. Trying to get an accurate reading on the postseason is always difficult, but there are some things we do know for certain.

1) Garnett put up similar box production in the postseason as he did in the regular season

As alluded to by Doctor MJ, here is Garnett RS --> PS

If we just look at the traditional box score for KG in '03-04:

RS: 24.2 PPG, 13.9 RPG, 5.0 APG 1.5 SPG 2.2 BPG
PS: 24.3 PPG, 14.6 RPG, 5.1 APG 1.3 SPG 2.3 BPG

RS: On +9.8, On-Off +20.7
PS: On +2.5, On-Off +26.7
His TS% drops from 54.7 to 51.3, and his TOV% rises from 10.5 to 14.9.

2) What were Garnett's opponents, and how did they stack up to Wade's?

The Bulls were clearly the worst team, the Nuggets and Nets were of similar caliber, the Kings were in a lot of ways similar to Detroit (and to a lesser extent Dallas, which was better than both but not by a large margin).

3) How did Wade and Garnett impact the game?

Well, let's look at how the team did with and without, how much lifting were these players doing?

Garnett On vs Off (% Change)
eFG% +.071
ORB% +2.7
DRB% +0.0
TRB% +3.2
AST% +20.0
STL% +0.2
BLK% +1.2
TOV% -5.0
ORtg +20.7

Wade On vs Off (% Change)

eFG% +.076
ORB% +11.6
DRB% +3.5
TRB% +6.8
AST% +0.9
STL% +2.4
BLK% -1.5
TOV% -1.9
ORtg +21.9

The numbers are eerily similar, almost as if both were having similar lifts offensively, with the largest differences in regard to offense being Offensive Rebounding, Assists, and Turnovers.

This information is interesting, but we have more data points that back up the notion that they were similarly impactful offensively.

Garnett's Offensive Impact in a nut-shell

2003: 5th in offensive win shares (OWS), 2nd in offensive PI-RAPM (ORAPM)
2004: 2nd in OWS, 1st in ORAPM

Ultimately, I have no reason to think Garnett played worse in the postseason. It doesn't show up on the tape, it doesn't show up in the box-score, and it doesn't show up in how Minnesota performed with or without Garnett, which mirrored the regular season trends. Nothing in the small sample of postseason is making me change my priors (notably 2003 and 2004).

[lessthanjake]

One thing I want to note about 2006 Wade is that a lot of that run came down to some very close games against great teams that the Heat should’ve lost to, and Wade was simply spectacular at the ends of these games.

For instance, in 92 minutes against the Pistons and Mavs in the 4th Quarter + OT with the score within 10 points, Wade averaged 50 points per 100 possessions, on a staggering 65.15% TS%. Remember, this is 2006, so that TS% is wild. Not coincidentally, the Heat had a +17.86 net rating in those minutes. And that basically was what won them the championship. Granted, 92 minutes is certainly a small sample, but I’m not talking about this to extrapolate anything about him as an overall player. A big part of greatness is specifically coming through in the most important moments. These were those moments, and Wade in 2006 came through to an incredible degree, such that he dragged a team to a title that really had no business winning one.

[ReggiesKnicks]

One thing I want to note about 2006 Wade is that a lot of that run came down to some very close games against great teams that the Heat should’ve lost to, and Wade was simply spectacular at the ends of these games.

For instance, in 92 minutes against the Pistons and Mavs in the 4th Quarter + OT with the score within 10 points, Wade averaged 50 points per 100 possessions, on a staggering 65.15% TS%. Remember, this is 2006, so that TS% is wild. Not coincidentally, the Heat had a +17.86 net rating in those minutes. And that basically was what won them the championship. Granted, 92 minutes is certainly a small sample, but I’m not talking about this to extrapolate anything about him as an overall player. A big part of greatness is specifically coming through in the most important moments. These were those moments, and Wade in 2006 came through to an incredible degree, such that he dragged a team to a title that really had no business winning one.

I don't hyper-fixate on tiny samples and over-index on minute samples, but I understand the appeal of the approach. It reminds me when ESPN used to be tolerable and not a Reality TV show and they would put up "Last 24 seconds or less" or "Last 5 seconds or less" as if those particular moments are more important than the 47 Minutes, 34 seconds prior to the 47 Minutes, 55 seconds prior.

I get the appeal, but it isn't what I value.

[lessthanjake]

Results-oriented thinking is valuable in the sense that results are important, but it can take away from actual player comparison and analysis.

I think results matter quite a lot in an assessment of the “greatness” of a player’s year, which is what this project is about. It shouldn’t be the only factor, but yeah I definitely think winning a title adds a lot to the “greatness” of a player’s year.

But also, results can tell us a good deal about how a player played. That the 2006 Heat had no business winning a title but yet did so despite Shaq diminishing before our very eyes definitely strongly suggests that Wade played amazingly well individually. Is it possible for that to have happened because the rest of the supporting cast played out of their minds rather than Wade playing amazingly well? Sure. But I watched these series (and can also look at the box score), and I know that what actually happened is Wade’s incredible individual play was what was driving this.

And even leaving aside the massive gap in what they led their teams to, at an individual level I think Wade was definitely better in the playoffs, and I certainly don’t think any series from Garnett in 2004 was as impressive as Wade’s series against the Pistons and Mavericks. Garnett was definitely a beast defensively, but he struggled with his scoring in those playoffs, putting up a negative rTS%.

Referencing rTS% for a player like Kevin Garnett as to why he struggled is showing a clear misunderstanding of a player like Kevin Garnett's ilk.

If we want to do a series-by-series analysis and analysis beyond struggled with his scoring in those playoffs as the leading measurement for Garnett's offense, we will see a clearer picture. Attempting to analyze offense through one lens, rTS%, isn't utilizing the important signals and tools we have at our disposal to properly assess these two all-time greats.

I think rTS% matters a whole lot when assessing a player who is taking like 24 true shot attempts a game in the playoffs. To say it “misunderstands” Garnett to talk about rTS% in reference to him is…very misguided. Is scoring efficiency his best attribute? Far from it. But things that aren’t someone’s best attribute still matter a lot, particularly if they’re doing those things with significant volume. Garnett took a lot of shots and he was much more efficient on them in the regular season than in the playoffs. Indeed, his scoring efficiency was actually a genuine negative in the playoffs.

You can say he brings other things to the table such that he can still have played well despite bad scoring efficiency. I’d be inclined to agree. But playing well is not the same as playing as well as he did in the regular season, nor is it the same as playing as well as 2006 Wade did in the playoffs. Those are high bars.

And it certainly doesn’t make sense to assert that Garnett played just as well in the playoffs as in the regular season and then when confronted with evidence that he did not do so, to just say that that evidence doesn’t matter because it doesn’t relate to his biggest strengths. It’d be like if Jokic played demonstrably worse defense in the playoffs than normal and someone said we shouldn’t care about that in assessing his playoffs because defense isn’t his strong suit. If a player actually did something, then how well he did it matters, and Garnett took a lot of shots in the 2004 playoffs and he did not make them efficiently.

You assert that Garnett was the better player in the playoffs, but without much reasoning for that beyond some reference to raw plus-minus data and a naked assertion that he “was just as good as he was in the regular season.” For reference, Garnett had a +2.9 EPM in the 2004 playoffs. That was not only a large drop off from the +6.8 he put up in the regular season, but was also miles below Wade’s +7.3 in the 2006 playoffs. Similarly, Garnett put up a 6.5 BPM in the 2004 playoffs, which was again a large drop off from the 10.2 he had in the regular season and far below the 9.3 that Wade had in the 2006 playoffs. Similarly, Garnett put up 0.163 WS/48 in the 2004 playoffs, way below his 0.272 in the regular season and Wade’s 0.240 in the 2006 playoffs.

So yeah, I think it’s pretty straightforward to conclude that Garnett was not as good in the 2004 playoffs as he was in the 2004 regular season, and that Wade was significantly better in the 2006 playoffs than Garnett was in the 2004 playoffs. The question is just whether Garnett being the more impactful regular season player overcomes that, for purposes of the “greatness” of the year. As I mentioned in my voting post, I initially started writing my voting post intending to argue exactly that, but then I realized while writing it that I didn’t buy it. Wade’s substantially better playoffs and the fact that he dragged a team to a title that I think had no business winning it simply weighs more highly to me.

I think it is important to remember that there is a lot of noise in all-in-ones for small sample sizes. What I attempt to do with small sample sizes is make them larger, which increases their accuracy by increasing the sample. Trying to get an accurate reading on the postseason is always difficult, but there are some things we do know for certain.

1) Garnett put up similar box production in the postseason as he did in the regular season

As alluded to by Doctor MJ, here is Garnett RS --> PS

If we just look at the traditional box score for KG in '03-04:

RS: 24.2 PPG, 13.9 RPG, 5.0 APG 1.5 SPG 2.2 BPG
PS: 24.3 PPG, 14.6 RPG, 5.1 APG 1.3 SPG 2.3 BPG

RS: On +9.8, On-Off +20.7
PS: On +2.5, On-Off +26.7
His TS% drops from 54.7 to 51.3, and his TOV% rises from 10.5 to 14.9.

2) What were Garnett's opponents, and how did they stack up to Wade's?

The Bulls were clearly the worst team, the Nuggets and Nets were of similar caliber, the Kings were in a lot of ways similar to Detroit (and to a lesser extent Dallas, which was better than both but not by a large margin).

3) How did Wade and Garnett impact the game?

Well, let's look at how the team did with and without, how much lifting were these players doing?

Garnett On vs Off (% Change)
eFG% +.071
ORB% +2.7
DRB% +0.0
TRB% +3.2
AST% +20.0
STL% +0.2
BLK% +1.2
TOV% -5.0
ORtg +20.7

Wade On vs Off (% Change)

eFG% +.076
ORB% +11.6
DRB% +3.5
TRB% +6.8
AST% +0.9
STL% +2.4
BLK% -1.5
TOV% -1.9
ORtg +21.9

The numbers are eerily similar, almost as if both were having similar lifts offensively, with the largest differences in regard to offense being Offensive Rebounding, Assists, and Turnovers.

This information is interesting, but we have more data points that back up the notion that they were similarly impactful offensively.

I find this response a bit odd. You basically say all-in-ones should be ignored here because of small sample sizes, but that we should instead look at playoff on-off information??? That’s obviously inconsistent, especially when playoff all-in-ones specifically aim to incorporate more data in order to stabilize extremely noisy playoff on-off data. Like, you’re dismissing one type of data on the grounds of low sample sizes, while simultaneously touting data that has an even bigger issue with noise due to low sample sizes. To put a point on this, your analysis is based on on-off data with an 86-minute off sample for Garnett. If trying to use that to assess Garnett’s impact, that’s obviously indexing on noise, while the data I provided is much more stable.

Ultimately, I have no reason to think Garnett played worse in the postseason. It doesn't show up on the tape, it doesn't show up in the box-score, and it doesn't show up in how Minnesota performed with or without Garnett, which mirrored the regular season trends. Nothing in the small sample of postseason is making me change my priors (notably 2003 and 2004).

You have no reason to think it if you ignore the ways that he was worse. He scored notably less efficiently. He turned the ball over notably more. And his volume on a per-possession basis across essentially all stats was lower (notice that his per-game stats were really similar despite playing 4 more MPG in the playoffs). So yeah, the dip in his quality of play definitely does “show up in the box-score.” And that is, of course, the main reason why he saw a very big dip in the various metrics I provided to you.

[ReggiesKnicks]

One thing I want to note about 2006 Wade is that a lot of that run came down to some very close games against great teams that the Heat should’ve lost to, and Wade was simply spectacular at the ends of these games.

For instance, in 92 minutes against the Pistons and Mavs in the 4th Quarter + OT with the score within 10 points, Wade averaged 50 points per 100 possessions, on a staggering 65.15% TS%. Remember, this is 2006, so that TS% is wild. Not coincidentally, the Heat had a +17.86 net rating in those minutes. And that basically was what won them the championship. Granted, 92 minutes is certainly a small sample, but I’m not talking about this to extrapolate anything about him as an overall player. A big part of greatness is specifically coming through in the most important moments. These were those moments, and Wade in 2006 came through to an incredible degree, such that he dragged a team to a title that really had no business winning one.

I don't hyper-fixate on tiny samples and over-index on minute samples, but I understand the appeal of the approach. It reminds me when ESPN used to be tolerable and not a Reality TV show and they would put up "Last 24 seconds or less" or "Last 5 seconds or less" as if those particular moments are more important than the 47 Minutes, 34 seconds prior to the 47 Minutes, 55 seconds prior.

I get the appeal, but it isn't what I value. I prefer nuance and large samples, but that's from the fascination for Data I have had for 50-some years.

[ReggiesKnicks]

I find it odd to see these two responses back to back. You say all-in-ones should be ignored here because of small sample sizes, and we should instead look at playoff on-off information???

I didn't say that!

What I am saying is we have these massive data samples, incredible in scope, spanning thousands and thousands of minutes. You and I can both agree that these large samples have strong indicators about a player, their strengths, and how they impact a game (notably how much).

Next, we have a smaller sample, a post-season. Not only is this sample small, but to make matters worse, the clear-cut 2nd-best player on the Timberwolves misses part of a pivotal series. When we look at the post-season sample, which is small, and the regular season sample (even including 2003 for an even larger sample, as KG is very similar in 2003 and 2004 as a player), we see a similar trend. Exceptional Box-Score, exceptionally simple +/-, massive shifts in the team's performance when Garnett is on the court versus when he is on the bench.

Now, we have a small sample size which is eerily similar to a large one. What do I say as a Data Analyst? That makes sense given the priors!

But EPM and other statistics aren't in love with Garnett like they were in the larger sample? Is it because Garnett was worse or because of the small sample?

I've watched the available tape on Garnett during his prime; he didn't look any worse in 2004. His reads were as sharp as ever, he was moving as a gazelle, defensively switching from Kobe to Shaq in the same possession, impressive stuff to say the least.

The team's performance is similar to his and is even more similar without him. All of this data I have says Garnett was just as good in the postseason as he was in the regular season, where he was on the cusp of a GOAT peak.

[lessthanjake]

I find it odd to see these two responses back to back. You say all-in-ones should be ignored here because of small sample sizes, and we should instead look at playoff on-off information???

I didn't say that!

What I am saying is we have these massive data samples, incredible in scope, spanning thousands and thousands of minutes. You and I can both agree that these large samples have strong indicators about a player, their strengths, and how they impact a game (notably how much).

Next, we have a smaller sample, a post-season. Not only is this sample small, but to make matters worse, the clear-cut 2nd-best player on the Timberwolves misses part of a pivotal series. When we look at the post-season sample, which is small, and the regular season sample (even including 2003 for an even larger sample, as KG is very similar in 2003 and 2004 as a player), we see a similar trend. Exceptional Box-Score, exceptionally simple +/-, massive shifts in the team's performance when Garnett is on the court versus when he is on the bench.

Now, we have a small sample size which is eerily similar to a large one. What do I say as a Data Analyst? That makes sense given the priors!

But EPM and other statistics aren't in love with Garnett like they were in the larger sample? Is it because Garnett was worse or because of the small sample?

EPM and other statistics aren’t in love with 2004 Garnett in the playoffs because his box data got notably worse. His output per possession went down essentially across the board, while his efficiency also went down. That gives good reason to think his impact went down as well, and data with a 86-minute off sample (a decent portion of which was likely garbage time) isn’t particularly good evidence to the contrary IMO.

I also should point out that there’s a bit of a logical fallacy here. We are discussing whether Garnett fell off in the playoffs compared to the regular season, and your reasoning for why he didn’t fall off in the playoffs is in part to look at “the regular season sample.” It seems obvious that regular season data is not proof either way about how he did in the playoffs compared to the regular season. You might use his great regular season data to say that even if he fell off in the playoffs he fell from such a high place that he was still great. But you cannot use great regular season data on its own to answer how he did in the playoffs compared to the regular season. To be fair, I get that your point is to say the on-off in the regular season and playoffs looks similar, but that’s just a roundabout way of relying on impact data in a tiny playoff sample. If the tiny playoff on-off sample is super noisy (which we both agree it is), then the fact that the numbers that got spit out happen to look at least somewhat similar to Garnett’s regular season on-off numbers is not particularly meaningful. With the amount of noise inherent in that on-off data, it basically amounts to relying on a random coincidence.

Finally, I note that EPM accounts for the same on-off data you’re looking at, since it has a RAPM component. Basically, in data that incorporates small-sample RAPM and box data, 2004 Garnett looks like he had a big fall off in the playoffs. Besides citing and comparing to regular season data (addressed above), your response is largely to cite to raw playoff on-off. Which is just similar but worse data than what is already incorporated into EPM (because RAPM is generally better than raw on-off), and EPM takes that and stabilizes it with additional information.

I've watched the available tape on Garnett during his prime; he didn't look any worse in 2004. His reads were as sharp as ever, he was moving as a gazelle, defensively switching from Kobe to Shaq in the same possession, impressive stuff to say the least.

The team's performance is similar to his and is even more similar without him. All of this data I have says Garnett was just as good in the postseason as he was in the regular season, where he was on the cusp of a GOAT peak.

You’re free to have your own view on the eye test, for sure. On the team performance front though, they did not do similarly well. Despite Garnett playing more MPG in the playoffs than in the regular season, the 2004 Timberwolves had a relative net rating of +3.8 in the playoffs, after having had a +6.1 net rating in the regular season. The team did worse in the playoffs, while Garnett’s volume and efficiency went down. It seems pretty consistent with Garnett having fallen off a significant amount in the playoffs.