Better peak: Kevin Garnett vs Nikola Jokic

[Peregrine01]

Not really. I stand by my take of being more impressed with Jokic. His work scoring the ball was wonderful against GSW, and something that translates in a vacuum. I think GSW, were a unique poor matchup for Jokic, and Denver still not having the best defensive infrastructure made him look as badly as possible.

I think the 03 and 04 losses, just made me feel that on offense, KG was a swiss-army, Jack of all trades, master of none (except passing); that passing he can't consistently leverage on offense, due to his not great first step and ability to drive to the rim. I leave those series more questioning, KG's true optimized role on offense, and how valuable that even is in this case.

Against the Lakers in 03 and 04, Minny scored on an ORTG of 105 in both series. This against a team that allowed 105 and 101 in the RS. The second highest scorers in each of those series? Troy Hudson at 24 ppg and Spreewell at 21 ppg.

Against the Warriors in 22, Nuggets scored on an ORTG of 115. This against a team that was first in the league, allowing 107 in the RS. The bare-boned Nuggets performed better against the Warriors defense than any team that the Warriors faced in their title run. The second highest scorer on the Nuggets in that series? Monte Morris at 14 ppg.

However better KG is on defense just doesn't make up the chasm between the two on offense.

Now look at DRTG as well...

Again, no one’s saying that the Nuggets didn’t get lit up. But what explains Minny getting lit up in the playoffs in 2002 or 2003? And Dirk, his matchup in 2002, completely torching him? Is it just because KG, the ATG defender, had bad defensive teammates? Or maybe he’s not as impactful as you think he is?

Anyway, feels like I’m arguing in circles with you. So I’ll just stop here.

[70sFan]

Against the Lakers in 03 and 04, Minny scored on an ORTG of 105 in both series. This against a team that allowed 105 and 101 in the RS. The second highest scorers in each of those series? Troy Hudson at 24 ppg and Spreewell at 21 ppg.

Against the Warriors in 22, Nuggets scored on an ORTG of 115. This against a team that was first in the league, allowing 107 in the RS. The bare-boned Nuggets performed better against the Warriors defense than any team that the Warriors faced in their title run. The second highest scorer on the Nuggets in that series? Monte Morris at 14 ppg.

However better KG is on defense just doesn't make up the chasm between the two on offense.

Now look at DRTG as well...

Again, no one’s saying that the Nuggets didn’t get lit up. But what explains Minny getting lit up in the playoffs in 2002 or 2003? And Dirk, his matchup in 2002, completely torching him? Is it just because KG, the ATG defender, had bad defensive teammates? Or maybe he’s not as impactful as you think he is?

Anyway, feels like I’m arguing in circles with you. So I’ll just stop here.

2002 was a bad showing from Garnett and his teammates, it's a clear black mark on his resume in my opinion. You see, I have no bias - Garnett played poorly in that series.

2003 isn't comparable to what Warriors did to Nuggets in 2022 and you know it.

[AleksandarN]

When was KG ever an elite offensive player?

Depending on your definition of "elite" (he was never Jokic on that end if you're asking this), but - I don't know, in 2003-04?

Best passing bigman of that era? I mean he was a decent passer, but would you have him over Divac for instance?

Arguably the best passer, yeah. You can make a case for Divac and that's basically it.

And however much value you want to ascribe to his passing, it just pales in comparison to his mediocre scoring efficiency and lack of ability to really bend defenses.

Garnett didn't have mediocre scoring efficiency at his peak, he was +3-3.5 percentage points above the league average in TS%.

Elite post isolation scorer? KG was often relegated to being a turnaround jump shooter from 15 feet out because of his lack of size in the post. He was hardly a guy that could really hurt you in isolation and teams were very comfortable letting him take those.

Tracking data we have suggest something different, at least for post up isolations.

Unfavorable matchups like the Suns were supposed to be in these past playoffs? And the Lakers as ewll?

Paul-less Suns and Lakers don't have guards who could torch you on spread P&Rs, so not really.

Only Curry's Warriors of last season and the CP3 Suns in 2021 really made the Nuggets' defense look bad but who was guarding those guys? Monte Morris, Will Barton and Facundo Campazzo? The shorthanded Nuggets also beat Dame's Blazers in 2021 despite that team being yet another prototype for exposing Jokic's defense.

Not only, the Jazz and Lakers in 2020 also abused their defense. That leads us to exactly one series against the Clippers from 2020 before 2023...

Also, the Nuggets won against Blazers while giving up 122 ORtg... How is that an argument for Jokic defense?

"Sub-20 wins" Sounds like you're pulling numbers out of your ass.

I am polite to you, why do you do that? We had this discussion during the top 100 project, you can find the numbers - and no, it's not my calculations.

But if you want to go there, the Nuggets without Jokic these last few years would've ranked dead last in the whole league in net rating.

I am aware of that, that's why I don't hold first round loss against Jokic, unlike what people do with Garnett.

The fact that you're willing to come up with so many excuses for KG's lack of team success until he joined up with a superteam but find ways to make excuses for why Jokic just won a championship reveals some notable bias.

Oh, so now I am biased against Jokic, even though he's my 3rd favorite active player and despite the fact that I'm Duncan fan who legitimately dislikes Garnett...

I have been called Jordan stan, LeBron stan, but now I am biased because I consider Garnett's case over Jokic. Sorry, but this is not a productive discussion. I always thought you are a good poster, but I won't engage with someone who accuses me of manipulating arguments for some imaginary biases.

I don't have any problem with someone picking Jokic here, I like him more as a player. I am just trying to be objective. I have nothing against people taking Garnett over Duncan either, even though I disagree. You shouldn't be so emotionally attached to your favorite player, it's only a basketball discussion.

CWebb was a way better passer than KG. What are we talking about. I would put Sabonis above him in that era

[Ryoga Hibiki]

Now look at DRTG as well...

Again, no one’s saying that the Nuggets didn’t get lit up. But what explains Minny getting lit up in the playoffs in 2002 or 2003? And Dirk, his matchup in 2002, completely torching him? Is it just because KG, the ATG defender, had bad defensive teammates? Or maybe he’s not as impactful as you think he is?

Anyway, feels like I’m arguing in circles with you. So I’ll just stop here.

2002 was a bad showing from Garnett and his teammates, it's a clear black mark on his resume in my opinion. You see, I have no bias - Garnett played poorly in that series.

2003 isn't comparable to what Warriors did to Nuggets in 2022 and you know it.

Actually I am quite convinced the 2003 TWolves without Garnett were quite better than the 2022 Nuggets without Jokic.

[70sFan]

Again, no one’s saying that the Nuggets didn’t get lit up. But what explains Minny getting lit up in the playoffs in 2002 or 2003? And Dirk, his matchup in 2002, completely torching him? Is it just because KG, the ATG defender, had bad defensive teammates? Or maybe he’s not as impactful as you think he is?

Anyway, feels like I’m arguing in circles with you. So I’ll just stop here.

2002 was a bad showing from Garnett and his teammates, it's a clear black mark on his resume in my opinion. You see, I have no bias - Garnett played poorly in that series.

2003 isn't comparable to what Warriors did to Nuggets in 2022 and you know it.

Actually I am quite convinced the 2003 TWolves without Garnett were quite better than the 2022 Nuggets without Jokic.

I know it doesn't tell the whole story and isn't a great indicator for full season, but:

2003 MIN: -18.2 without Garnett
2022 DEN: -7.7 without Jokic

Why do you that's the case?

[nate33]

2002 was a bad showing from Garnett and his teammates, it's a clear black mark on his resume in my opinion. You see, I have no bias - Garnett played poorly in that series.

2003 isn't comparable to what Warriors did to Nuggets in 2022 and you know it.

Actually I am quite convinced the 2003 TWolves without Garnett were quite better than the 2022 Nuggets without Jokic.

I know it doesn't tell the whole story and isn't a great indicator for full season, but:

2003 MIN: -18.2 without Garnett
2022 DEN: -7.7 without Jokic

Why do you that's the case?

Garnett averaged 40.5 minutes a game over 82 games. There were only 615 non-Garnett minutes, many of which were presumably garbage time. The "without Garnett" minutes sample size was too small to be very accurate.

[70sFan]

Actually I am quite convinced the 2003 TWolves without Garnett were quite better than the 2022 Nuggets without Jokic.

I know it doesn't tell the whole story and isn't a great indicator for full season, but:

2003 MIN: -18.2 without Garnett
2022 DEN: -7.7 without Jokic

Why do you that's the case?

Garnett averaged 40.5 minutes a game over 82 games. There were only 615 non-Garnett minutes, many of which were presumably garbage time. The "without Garnett" minutes sample size was too small to be very accurate.

I know that, but I wonder what convinced him. What other evidences do we have?

[Woodsanity]

My problem with KG was that ultimately he was a 52 TS% guy in the playoffs.
He was a great passer but was an overrated scorer.

Elite offense has always been valued over elite defense.

I will take Jokic whose defense has improved enough to not be a liability in the playoffs.

[homecourtloss]

Jokic’s defensive ON/OFF DRAPTOR doesn’t look good, whereas the box defensive RAPTOR does. He’s an elite defensive rebounder but as such, you’d think that being so would translate to the plus/minus numbers we have but generally doesn’t. Also, whatever other box aspects RAPTOR likes about Jokic’s defense doesn’t seem to translate to plus/minus numbers.

I don’t understand this desire to use a *component* of RAPTOR as if it’s an independent stat. It’s not. There’s very good reason that that’s just a component of the stat, and it’s literally because the overall stat is substantially more accurate. So this is just you saying “Yeah, a stat says Jokic’s defense is great, but one subpart of that stat that is obviously less accurate than the overall stat doesn’t think Jokic’s defense is good.” Not exactly a persuasive point. You should also note that the “box” portion of RAPTOR uses tracking data—it’s not just taking numbers from the box score.

First of all, your use of “Obviously less accurate”— what does “accurate” mean here? Do we know the true goodness of a player’s defense to compare the “accuracy” between BOX and ON/OFF? If so, where does that true goodness exist and who has access to it? How it is obvious if we don’t know the true goodness of the player’s defense?

Secondly, we have information about elements that have known to affect defense in certain ways, but that information isn’t new as THAT Information already exists (tracking data, box elements, etc.). We already have access to these things so I’m not sure how getting a box number with subjective weights assigned increases our knowledge of accuracy. The only thing new is how 538 weighs certain elements that they don’t seem to disclose. I don’t need 538’s specific number that weights certain elements in whatever way it wants as it doesn’t tell us something we didn’t already know. I can look at box components separately as well as tracking elements—I don’t need a BOX element that assigns whatever elements. I don’t know what weights what elements are getting to create the Box DRAPTOR, e.g., DREB%, STL%, BLK%, shot defense tracking, rim protection, height adjustments (something you called weird when discussing some of JE’s work), etc.

Additionally, I’m not sure why the Box DRAPTOR portion gets so much more weight than the ON/OFF DRAPTOR (also don’t know how this is calculated).

Box DRAPTOR and Overall DRAPTOR had an r of .971



ON/OFF DRAPTOR and overall DRAPTOR had an r of .67



Seems odd that the box element gets this much weight—the difference between the overall DRAPTOR and the box raptor is basically the same (avg. difference is like .02) while the avg. difference between On/OFF DRAPTOR and overall DRAPTOR is 6 times higher. I don’t know about you, but the sizable differences makes me want to break down the components.

[lessthanjake]

Additionally, I’m not sure why the Box DRAPTOR portion gets so much more weight than the ON/OFF DRAPTOR (also don’t know how this is calculated).

Well, they did provide an explanation of that: https://fivethirtyeight.com/features/introducing-raptor-our-new-metric-for-the-modern-nba/

Spoiler:

Overall, however, RAPTOR weights the “box” component more highly than the “on-off” component. In testing RAPTOR on out-of-sample data, we found that while on-court/off-court stats provide useful information, they’re nonetheless quite noisy as compared with individual measures of player value that are used in the “box” part of RAPTOR.

Basically, they used the weighting they used because it tested better on out-of-sample RAPM data. In other words, it seemed to make the measure more accurate.

[homecourtloss]

Additionally, I’m not sure why the Box DRAPTOR portion gets so much more weight than the ON/OFF DRAPTOR (also don’t know how this is calculated).

Well, they did provide an explanation of that: https://fivethirtyeight.com/features/introducing-raptor-our-new-metric-for-the-modern-nba/

Spoiler:

Overall, however, RAPTOR weights the “box” component more highly than the “on-off” component. In testing RAPTOR on out-of-sample data, we found that while on-court/off-court stats provide useful information, they’re nonetheless quite noisy as compared with individual measures of player value that are used in the “box” part of RAPTOR.

Basically, they used the weighting they used because it tested better on out-of-sample RAPM data. In other words, it seemed to make the measure more accurate.

Interesting that they would be doing that because if you look at Nikola Jokić, for example, his on/off RAPTOR rankings correspond strongly with single-season NBA shotcharts RAPM, whereas his box DRAPTOR doesn’t and overall DRAPTOR looks like basically just the box DRAPTOR number anyway as seen in the previous post.

If you took every season from 2013-2014 to this past season and ranked all the rotation level players’ respective box DRAPTOR numbers, Jokic has the 5th, 26th and 37th highest box DRAPTORs out of 2,440 player seasons. His on/off DRAPTOR highest are 125th, 125th and 676.

Meanwhile, if you took all the seasons from 2013-2014 to this past season and looked at the single-season DRAPM they are around 130th, 180th, and 750th. . Also, on/off DRAPTOR and DRAPM strongly correspond with each other, while box DRAPTOR and DRAPM don’t.

[lessthanjake]

Additionally, I’m not sure why the Box DRAPTOR portion gets so much more weight than the ON/OFF DRAPTOR (also don’t know how this is calculated).

Well, they did provide an explanation of that: https://fivethirtyeight.com/features/introducing-raptor-our-new-metric-for-the-modern-nba/

Spoiler:

Overall, however, RAPTOR weights the “box” component more highly than the “on-off” component. In testing RAPTOR on out-of-sample data, we found that while on-court/off-court stats provide useful information, they’re nonetheless quite noisy as compared with individual measures of player value that are used in the “box” part of RAPTOR.

Basically, they used the weighting they used because it tested better on out-of-sample RAPM data. In other words, it seemed to make the measure more accurate.

Interesting that they would be doing that because if you look at Nikola Jokić, for example, his on/off RAPTOR rankings correspond strongly with single-season NBA shotcharts RAPM, whereas his box DRAPTOR doesn’t and overall DRAPTOR looks like basically just the box DRAPTOR number anyway as seen in the previous post.

If you took every season from 2013-2014 to this past season and ranked all the rotation level players’ respective box DRAPTOR numbers, Jokic has the 5th, 26th and 37th highest box DRAPTORs out of 2,440 player seasons. His on/off DRAPTOR highest are 125th, 125th and 676.

Meanwhile, if you took all the seasons from 2013-2014 to this past season and looked at the single-season DRAPM they are around 130th, 180th, and 750th. . Also, on/off DRAPTOR and DRAPM strongly correspond with each other, while box DRAPTOR and DRAPM don’t.

It’s trying to match with RAPM data in the aggregate. The whole point of adding box components is that pure RAPM is really noisy in smaller samples (which single seasons certainly applies to, but even longer timeframes than that for a given player are noisy), and you can improve it in smaller samples by adding more data.

[70sFan]

How DRAPTOR was tested exactly? Where can I find the info?

[Pelly24]

Going Jokic. I think unless you end up with an improbably well-balanced team, you won't win a chip without a transcendent offensive force like an MJ, Magic, LeBron, Bird, Kobe, Luka, Hakeem, Curry, Wade... etc. I think you can build a defense that can almost approximate a KG effect, but if you don't have one of the types of players I list above, it's going to be very hard to win a chip. So I give it to Jokic. He just averaged a 30-point triple double on 63 TS% for the whole playoffs. They only lost four games the whole playoffs, going through KD, Booker, Ant Edwards and KAT, (injured) LeBron and AD and then (banged up) Jimmy and Bam. His all-around dominance was the kind I haven't seen since 2018 LeBron. It was just inevitable that he was going to do whatever he wanted to do at the end of the day

[70sFan]

Going Jokic. I think unless you end up with an improbably well-balanced team, you won't win a chip without a transcendent offensive force like an MJ, Magic, LeBron, Bird, Kobe, Luka, Hakeem, Curry, Wade... etc. I think you can build a defense that can almost approximate a KG effect, but if you don't have one of the types of players I list above, it's going to be very hard to win a chip. So I give it to Jokic. He just averaged a 30-point triple double on 63 TS% for the whole playoffs. They only lost four games the whole playoffs, going through KD, Booker, Ant Edwards and KAT, (injured) LeBron and AD and then (banged up) Jimmy and Bam. His all-around dominance was the kind I haven't seen since 2018 LeBron. It was just inevitable that he was going to do whatever he wanted to do at the end of the day

You think teams like 2003-07 Spurs, 2008 Celtics, 2021 Bucks are improbably well-balanced teams?

By the way, putting Hakeem in that tier is questionable to say the least.

I also wouldn't say you can build a KG-led defense around Jokic.

[Jaivl]

Going Jokic. I think unless you end up with an improbably well-balanced team, you won't win a chip without a transcendent offensive force like an MJ, Magic, LeBron, Bird, Kobe, Luka, Hakeem, Curry, Wade... etc. I think you can build a defense that can almost approximate a KG effect, but if you don't have one of the types of players I list above, it's going to be very hard to win a chip. So I give it to Jokic. He just averaged a 30-point triple double on 63 TS% for the whole playoffs. They only lost four games the whole playoffs, going through KD, Booker, Ant Edwards and KAT, (injured) LeBron and AD and then (banged up) Jimmy and Bam. His all-around dominance was the kind I haven't seen since 2018 LeBron. It was just inevitable that he was going to do whatever he wanted to do at the end of the day

You think teams like 2003-07 Spurs, 2008 Celtics, 2021 Bucks are improbably well-balanced teams?

By the way, putting Hakeem in that tier is questionable to say the least.

I also wouldn't say you can build a KG-led defense around Jokic.

Always funny to talk about balanced teams and then pop up the usual 80s-90s names, when Basketball Was Right (TM), like Magic or Jordan didn't have extremely loaded +50 win supporting casts themselves or like Bird wasn't surrounded by 4-5 literal all-stars, FMVPs and MVPs most years.

[wafflzgod]

Hmm I have them super close

My Evals of them have them basically as a wash from a single season perspective
23 Jokic: 6 Off / -0.13 Def / +2 port --> 26.6
04 KG: 2.75 Off / 3.37 Def / +2 port --> 26.5

I might lean Jokic slightly just because I tend to prefer outlying offensive forces over two-way players if their impact is at a similar level. Could definitely be swayed though

[70sFan]

Hmm I have them super close

My Evals of them have them basically as a wash from a single season perspective
23 Jokic: 6 Off / -0.13 Def / +2 port --> 26.6
04 KG: 2.75 Off / 3.37 Def / +2 port --> 26.5

I might lean Jokic slightly just because I tend to prefer outlying offensive forces over two-way players if their impact is at a similar level. Could definitely be swayed though

That's interesting numbers, would you be kind to share your evaluation of all top 12 post-1980 peaks (also pre-1980 if you have them)?

[wafflzgod]

Hmm I have them super close

My Evals of them have them basically as a wash from a single season perspective
23 Jokic: 6 Off / -0.13 Def / +2 port --> 26.6
04 KG: 2.75 Off / 3.37 Def / +2 port --> 26.5

I might lean Jokic slightly just because I tend to prefer outlying offensive forces over two-way players if their impact is at a similar level. Could definitely be swayed though

That's interesting numbers, would you be kind to share your evaluation of all top 12 post-1980 peaks (also pre-1980 if you have them)?

Top 12 Post-1980 for me currently
12 Lebron: 6 Off / 2 Def / -1 port --> 32.6
91 MJ: 6.37 Off / 1.13 Def / 0.5 port --> 32.1
00 Shaq: 5.5 Off / 1.5 Def / 1 port --> 30.3
93 Hakeem: 3.5 Off / 3.63 Def / -1 port --> 29.2
16 Steph: 6.13 Off / 0.13 Def / 2 port --> 28.3
86 Bird: 5.5 Off / 0.75 Def / 2 port --> 28.0
03 Duncan: 3.25 Off / 3.37 Def / 0 port --> 27.6
87 Magic: 6.5 Off / 0.13 Def / -0.5 port --> 27.1
23 Jokic: 6 Off / -0.13 Def / 2 port --> 26.6
04 KG: 2.75 Off / 3.37 Def / 2 port --> 26.5
22 Giannis: 3.5 Off / 2.75 Def / -0.5 port --> 25.7
08 Kobe: 5.63 Off / 0.5 Def / 0 port --> 25.5

[homecourtloss]

Hmm I have them super close

My Evals of them have them basically as a wash from a single season perspective
23 Jokic: 6 Off / -0.13 Def / +2 port --> 26.6
04 KG: 2.75 Off / 3.37 Def / +2 port --> 26.5

I might lean Jokic slightly just because I tend to prefer outlying offensive forces over two-way players if their impact is at a similar level. Could definitely be swayed though

That's interesting numbers, would you be kind to share your evaluation of all top 12 post-1980 peaks (also pre-1980 if you have them)?

Top 12 Post-1980 for me currently
12 Lebron: 6 Off / 2 Def / -1 port --> 32.6
91 MJ: 6.37 Off / 1.13 Def / 0.5 port --> 32.1
00 Shaq: 5.5 Off / 1.5 Def / 1 port --> 30.3
93 Hakeem: 3.5 Off / 3.63 Def / -1 port --> 29.2
16 Steph: 6.13 Off / 0.13 Def / 2 port --> 28.3
86 Bird: 5.5 Off / 0.75 Def / 2 port --> 28.0
03 Duncan: 3.25 Off / 3.37 Def / 0 port --> 27.6
87 Magic: 6.5 Off / 0.13 Def / -0.5 port --> 27.1
23 Jokic: 6 Off / -0.13 Def / 2 port --> 26.6
04 KG: 2.75 Off / 3.37 Def / 2 port --> 26.5
22 Giannis: 3.5 Off / 2.75 Def / -0.5 port --> 25.7
08 Kobe: 5.63 Off / 0.5 Def / 0 port --> 25.5

Interesting. How did you get to these numbers? Also, are these the entire season or just post season?