ardee wrote:I am afraid you will have to do better than this wondrous "invisible impact". As I have said before, the entire conventional top 15 were amazing players and you don't need to dig too deep to realize it. Why are we supposed to just take it for granted that even though it's not visible for Garnett, we're just supposed to take the KG arguers' word for it?
70sFan wrote:It's the first time in my life I see someone say that Timmy's only advantage over KG offensively is his PS scoring. Now it's like KG is the king of non-boxscore offense but Duncan didn't have much impact outside of scoring?
...
I don't want to touch defense because I also think that Timmy is better here, but that's not my point. My point is that most KG fans assume that Garnett has higher impact in all non-boxscores aspect of offense than any other bigman in NBA history (maybe besides Shaq). He's the only player who is praised for that. Meanwhile, Moses never get enough credit for his gravity factor. Or Robinson for his GOAT faceup game. I don't even say that KG is worse than Moses or Admiral, it's just my way to show the subject.
Actually, I really appreciate that KG fans find these "small things" important. All basketball fans should do that. I'm huge Wes Unseld fan because of that. But KG fans are focused on other players weaknesses a bit too much. Garnett himself wasn't exactly flawless. In 2008 when he was still close to his prime self, he still wasn't stunning scorer in playoffs.
andrewww wrote:To be it bluntly, Odom was the 2nd best player on LAL while Wally as on MIN. Everyone else on both teams was nothing to right about. KG actually had better scoring around him than Kobe did. So did Kobe's offense do more for LAL than KG's defense did for MIN? The results are the lakers made the playoffs came to within 1 game up 3-2 of upsetting MVP Nash's Suns. KG's team missed the playoffs in consecutive years.
It's interesting. In the 2014 project, there was a big push back against RAPM as a measure. The argument was that, similar to Outsider's post last thread, one shouldn't use a stat/approach that wasn't intuitive to many posters, and that gave results people didn't agree with. Instead, there should be more discussion of scouting, telling WHY a player makes an impact, than just relying on numbers.
I feel like, in this project, for the most part people have tried to do more of the scouting-based arguments, or mixing the scouting more with the numbers. I think this is a step in the right direction, as a whole. But an unintended consequence seems to be the type of push-back quoted above, where people act like the opinions expressed in the scouting-based posts have NO quantitative support. Like, the "KG fans", as they put it, have this strange opinion that his supporting casts were bad or that he was making a huge impact with them...but that there's nothing to back it up.
This, obviously, isn't the case. Side note, but what's ALSO not the case is, most of the KG supporters in this project are NOT KG fans. I am. I believe that Colbini may be? But Doc MJ isn't. Therealbig3 isn't (he was more of a Duncan fan). Not sure about MicahClay...but I think he's also a Duncan fan. The point is, this isn't a group KG love fest with an agenda to get their favorite player in. On the contrary, most of the KG support on this board is more like Doc MJ, that followed their analysis and came to their conclusions, then supports those conclusions even in the face of opposing viewpoints.
And, more focused, there is clearly a lot of data to support the viewpoints (here) that Garnett's casts actually were worse than, for example, 06/07 Kobe's, and it can be shown clearly without relying just on opposing opinions. Similarly, Garnett's impact is not at all invisible or only the word of the "KG fans"...it, too, can be shown very quantitatively.
1) Supporting casts quantified. In an earlier post,I broke down the casts of the 2006 and 2007 Timberwolves, Lakers and Mavericks as a response to someone suggesting that there wasn't all that much difference between what KG, Kobe and Dirk had to work with those seasons since none really had star-studded support. I used 3 different boxscore methods (summed Win Shares, summed Wins Produced, summed VORP among the teammates) and the "off-court" portion of the on/off +/- for all three teams to show the clear quantitative pattern that the 06 and 07 Timberwolves support was on an entirely different level of stink than what Kobe or Dirk was facing.
But, posts can get buried in this type of long project, and as we saw today, people will join the debate at different points and completely ignore things that have been said before. Plus, as Outsiders alluded to, sometimes my posts are wall-of-text and can have lots of numbers. So, thought I'd try something new here as a way to present numbers that are supporting previously stated scouting/qualitative arguments: graphs. They're visual, quick, and can be inserted in a post without having to keep track of all the numbers. So, below, find 2 graphs: 1) the average of total Win Shares and Wins Produced for all Spurs and Timberwolves teams (excluding KG and Duncan) from 1998 - 2007 with data added for 05 - 07 Kobe & 06 - 07 Dirk, 2) the sum of the VORP of all Spurs, Wolves, Lakers and Mavs squads from 1998 - 2007 (Dirk's starts in 2001).


To be clear, these plots come purely from boxscore results. I'm on record that the boxscores emphatically don't cover the whole game, and are particularly bad for defense. So, I consider these estimates to be reasonable ball-parks, but if anything they tend to not do a great job pegging defensive production. Nevertheless, their message is exceedingly clear. And they show, very clearly, that the "Garnett had poor supporting casts" line of reasoning isn't just hand waving, isn't just the subjective opinion on "KG fans", and extends WELL before the 2006 & 2007 crater periods.
No, Garnett's casts were really, quantitatively poor relative to his peers for the entirety of his career in Minnesota.
*Even the famed 2004 cast of Cassell and Sprewell was producing SIGNIFICANTLY less than an average year of support from his peers.
*06 Kobe's Lakers cast (that he led to 45 wins) was NOT comparable to 06 KG's cast...but was instead more comparable to the 2000 - 2002 Wolves casts that KG led to two 50-win seasons and one 47-win season.
*07 Kobe's Lakers cast that he led to 42 wins was, again, was NOT comparable to 07 KG's cast...but was instead more comparable to the 2003 Wolves cast that KG led to 51 wins.
And again...these are purely boxscore-based estimates that don't fully encompass how terrible the defensive talent was on those Wolves, that doesn't capture how much many of those Wolves teams depended on KG's ability to create offense, and that doesn't even TOUCH on the dreaded +/- stats that do help to illustrate these things. In fact, about those +/- numbers...
2) Garnett's very visible impact, quantified. Of course, there have been iterations of +/- studies through the years. And from the days that 82games.com first started publishing on/off +/- publicly, Garnett has consistently dominated each of these studies. And he continues to dominate them, even and especially once the datasets get big enough that the noisy, fluke-type situations wash out.
Again, this is just a numbers reporting post to support previous scouting-based posts. So, I'm not even going to try to put these in super context, or to explain the differences in the studies, or really do anything here besides report publicly available analytics conclusions that were NOT made up by "KG fans":
*
DocMJ's spreadsheet of publicly gathered single-year PI-RAPM from 1999 - 2016, normalized/scaled (I modified to include 2013 - 2016, 2002 full-season, dropped off 1998 because of reported error); best 5-years, averaged:
1) LeBron: +11.6
2) Garnett: +10.8
3) O'Neal: +9.5
3b) Duncan: +9.5
5) Nowitzki: +9.3
*02 - 11 RAPM study https://sites.google.com/site/rapmstats/10-year-rapm1) LeBron 8.8
2) Garnett 8.0
3) Wade 6.2
4) Manu 6.1
5) Kobe 6.1
6) Paul 5.9
7) Duncan 5.8
8) Nash 5.7
9) Dirk 5.6
*97 - 14 RAPM study https://sites.google.com/site/rapmstats/97-14-rapm-21) LeBron 5.6
2) Duncan 5.1
3) Shaq 5.1
4) Manu 5.1
5) Garnett 4.8
6) Robinson 4.5
7) Wade 4.5
8) Dirk 4.2
15-year RAPM study w/ no age curves:
https://docs.google.com/spreadsheets/d/1CA4KxmzjZrTlYqxNU85jkUnCcqvJjsP5LT818LSYjkk/edit#gid=01) LeBron 9.3
2) Garnett 9.1
3) Paul 7.6
4) Duncan 7.3
5) Dirk 6.9
I'll stop there. The numbers kind of speak for themselves. No matter who did a particular study, what methodology they used, what time span, Garnett always comes in right at the top, comparable to anyone else that's played in the last 20 years. These numbers didn't come from the imagination of KG fans. Also, those supporting KG in this project didn't just pull the concept of his huge impact on his team's results out of the air.
With scouting-based posts, several people in this project have tried to point out the mechanisms for why he makes such a monstrous difference on the court...and why so much of what he does isn't captured by the boxscores. Help defense, pick-and-roll defense, vocally coordinating the defense as a "middle linebacker", team offense initiation/creation, jump-shot spacing, mega gravity...these are all REASONS for why he's such a game changer that aren't in the boxscores. But the results of his contributions, and how they improve his teams...those are very much quantifiable.
This isn't hand waving. It's not a fluky, noisy result. Garnett IS one of the biggest impact players in NBA history, and his non-boxscore contributions ARE that big. You may not choose to value him in that way, but you also can't consign this to the imaginations of "KG fans". His impact is extremely visible, if you care to look.
Vote: Kevin Garnett
2nd: Shaquille O'Neal