GilmoreFan wrote:You're still pushing what is basically a made up stat over results and observation, and what you said to Richboy, that "the arguments that you make could very easily and reasonably be reversed and used against the position that you are defending" is far more true of the positions I have seen you take. You're obviously a huge KG fan, and every single position I've seen you take is in furtherance of propping him up in some way. Every possible argument or position which could break both ways, and even some that don't, you push in a way that favours KG. I saw you call T.Brandon a "solid PG" among other factually dubious claims, just to play down his cast, ignoring the same advanced stats that suggested Brandon outplayed KG one playoffs, when he shot 385. from the field.. You have an excuse for why KG failed every single year from 97-2003 and 2005-7. I'm willing to concede that KG's teams sucked in 2003 and 2007, and I even think that the correct choice between KG and Dirk has to be KG, but you do not give the impression of objectivity. There is no new evidence or stat that could be presented to you that would make you change your mind, in fact you thought the same thing before APM or other advanced stats were really used, so in effect you're going off the "eye test" too.
Lot of truth in what you wrote here, though some not-so-truth as well.
1) Yes, I am a KG fan.
2) Yes, I made my decisions about how well KG was or wasn't playing based on the "eye test" and not statistics
3) Clarification: I think you misunderstood what I said about the "eye-test" in my last post. It's an important part of the analysis process, maybe the most important. But where it's weaker is when it's used as arguments on its own, especially if it's in the contradiction to other available information and doesn't adapt. So certainly, use your "eye test" to form your own opinion. But don't later try to tell me that your opinion is better than mine or someone else's...you're going to have to use objective analysis to be convincing of that.
4) T. Brandon in Minnesota was a "solid PG". He was never among the top 25 players in the NBA at any point in his tenure. He was a good player, as I've said, probably the best player that KG played with in Minnesota outside of the 1 year of Cassell. But he did not stand out as a 2nd option compared to the David Robinson/Kobe Bryant types in San Antonio and LA, and he was not enough to bridge the gap in depth of talent that the Blazers, Kings and Mavs had during that time period. You can characterize Brandon however you like, if "solid" and "best teammate KG had" isn't to your liking. But factually speaking, Brandon and the rest of the Wolves were a marked step down from the casts of the other Western powers. Which, ultimately, is the point.
5) Back to my own decision-making. As you pointed out, I was of the belief that KG was at least as good as Duncan from watching both of them play a lot. But why? I'm not from Minnesota and I had no pre-Wolves ties to KG. When I got League Pass in 1999 I kept both the Wolves and the Spurs in steady rotation, and after some time I came to the conclusion that I liked how Garnett played and that he was at least as good as his #21 counterpart. So while yeah, I'm likely biased now, I wasn't always. I came into it neutral, and became "biased" based on what I observed.
6) Decision-making continued: John Hollinger, Kevin Pelton, Justin Kubatko and Dave Berri are not, to my knowledge, Kevin Garnett fans. They are four guys that all, independently, came up with ways to analyze the NBA box scores to try to estimate how good individual players are. All of their methods have flaws, no doubt, but when all 4 methods all independently say that Garnett measures out among the top-2 (if not #1 outright) in the NBA over a full decade then yeah, I give that some credence. As we've talked about, my own "eyeball test" at some point became less objective once I became a Garnett fan. So when not just one, but every major boxscore-based advanced stat comes to the same conclusion as I do over such a long period of time, then yeah, I take notice of it.
7) Decision-making part 3: There are those that say that box score stats, no matter how "advanced", don't really matter because of "stat padding", especially on poorer teams. Fair enough. I certainly wouldn't base the entirety of my analysis on what the box scores tell me, any more than I would rely only on accolades or only on team results or only on the "eye test". I think you've got to use everything. So, that said, the family of +/- stats, to me, is another break through in that it gives us another objective way to examine players that is completely independent of all of the others. This family of stats cares NOTHING about the box scores...it is entirely based upon how an individual player is able to affect the team's results. And again, those that have developed this research like Wayne Winston, Stephen Ilardi, Dan Rosenbaum and Jeremias Englemann (among others) are not in any way associated with KG or the Wolves. So when, again, every major +/- based advanced stat comes to the same conclusion as I did based on watching which was the same as what others reached by examining the box scores...yeah, to me that adds even further credence.
8) The test. As Richboy pointed out, one of the real tests for any type of analysis is the question of predictive value. All of the above that I wrote about was true in the summer of 2007. By that summer I thought that Garnett was as good as any player in the NBA, there were already boxscore-based advanced stats saying the same, and the fledgling +/- stats were also suggesting the same thing. But outside of the egg-head community, the narrative was at this point firmly established that Duncan had moved beyond KG, and the fact that the Wolves were missing the playoffs every year was the nail in the coffin for KG as ultra-elite. So when KG was traded to Boston to play with Pierce and Allen, we had a nice test situation set up.
The majority opinion of those '08 Celtics was that they would be good, not great. ESPN's crew predicted they'd probably win the Atlantic division, but on average they picked them to finish 3rd in the East ( http://sports.espn.go.com/nba/dailydime ... review0708 ). Not a single one of them picked the Celtics to actually win the title ( http://sports.espn.go.com/nba/preview20 ... -NBAChamps ).
On the other hand, I was all-in that the Celtics would win that title. To me, this was like a steroid-version of the '04 Wolves. KG with 2 strong perimeter players and a squad full of role players outside of that? As soon as the trade was announced I called a friend of mine in LA that often goes to Vegas, and had him put some money down on the Celtics to win it. It was a no-brainer to me. If what I and the various analysis methods that I used suggested was actually true and KG really was the best player in the NBA, then even slightly past his prime there was no way you could give him a cast that good and not have it result in a title.
But at the time, that WAS a true test because it was very much a minority view that those Celtics would win. In hindsight we have people talking about them having a "stacked" line-up or like it was a foregone conclusion that they'd win. But real-time, that was absolutely not the case. And here, again, I got more support for my previously held opinion. The Celtics blew through the season, won the title, KG was again among the league leaders in every advanced stat and was by-far the leader on the Celtics in all of them, and the results looked exactly like what you might have expected if your hypothesis being tested was that Garnett was the best player in the NBA.
Conclusion: so yeah, while I think both you and Richboy often make solid contributions, I don't agree with your conclusions. Just like you don't agree with mine. But disagreement is fine. It's even necessary on a message board. But to characterize my side of the agreement as disconnected with reality is absurd. You feel like you've got some evidence on your side, and that's fine. But I know I've got a bunch on my side as well. We can agree to disagree, but of a certainty, your disagreement with me doesn't in any way make my position less valid than yours.
That long post was powerful Drza, powerful.
