Garnett VS Nowitzki-Year By Year Since 2001

[Texas Chuck]

yeah I deleted that part because it was irrelevant.

I brought up poker because only bad/losing players bring up luck/variance. Everyone has to deal with luck/variance but play enough hands and that evens out and the best players win. Just like over 82 games the luck/variance evens out and you have to go by what actually happens.

[Texas Chuck]

But again this discussion is beyond pointless--you cant/wont see past statistics and I dont believe in statistics that tell me one team should have won more games than the team that actually did. Lies, Damn Lies, and Statistics.

Maybe Im too old-school (or just plain old) but I value what actually happens more than some computer-simulation of what might happen if we played the season out 100(1000?, 10,000?) times.

[Ginobili]

Well, mystic is the same guy that used his magical statistics to support the idea of "Deng being the MVP of the Bulls".

We all know what happened in the Playoffs with the Bulls.

[mysticbb]

Just like over 82 games the luck/variance evens out and you have to go by what actually happens.

If that would be true, record would be the best predictor for playoff success. But it isn't. SRS is one tool which is better. So, the luck/variance does not "even out" within 82 games. Let alone that Garnett was out for 11 games in 2008, which might have an effect on this.
The Mavericks in 2007 won more games than they should have won according to their scoring margin. They went 12-3 in close games, something over the expected rate in close games for teams with the respective strength of the Mavericks, the Celtics went 7-6, below the expected rate. The matter of fact is that you are wrong about "luck/variance evens out", because 82 games is not a big enough sample. Just like in Poker 82 games are not enough in order to even out luck/variance.

[MisterWestside]

Well, mystic is the same guy that used his magical statistics to support the idea of "Deng being the MVP of the Bulls".

We all know what happened in the Playoffs with the Bulls.

"But, but, RAPM *said* he's the MVP!"

[Texas Chuck]

no 82 games isnt close in poker but in basketball its all any one team ever gets so it has to do. Any statistical analysis that wont accept that premise is even more flawed than the silly per 36 stat that has grown in popularity in recent years.

[mysticbb]

Now that I bring up the 10-yr RAPM against your favorite player is when you decide to go over the "flaws" with the dataset :lol:

That is a great, Texas Chuck thinks I'm a Garnett/Celtics homer and you think I'm a Nowitzki/Mavericks homer. Well, I probably have a dissociative identity disorder, but maybe you, being such a great psychologist can answer that for me better ...

The issue I brought up is not just related to RAPM, but a problem with all kinds of things. I explained it once with superposition of long and short waves. If the timespan is short enough, you will not notice the longer wave, but when you increase the timespan, you will end up seeing amplifications and reductions. A 10yr sample shows exactly that problem, you have multiple players which actually end up with a higher 10yr RAPM than they ever had during the timespan when looking at the prior informed values. In the same sense you see players with much lower 10yr RAPM values than those they had during any given year in terms of prior informed RAPM values. But well, you may as well just try to ignore that.

[MisterWestside]

The issue I brought up is not just related to RAPM, but a problem with all kinds of things.

And you still missed the point of that post.

[mysticbb]

And you still missed the point of that post.

Well, for sure, if you say so. It is obvious that you are great enough as a psychologist to know exactly what kind of reaction was caused by you "exposing me". :lol:

Seriously, you are confusing things here. I didn't use raw +/-, nor did I use 1yr RAPM. The first number was adjusted for 100 possessions, the next was based on prior informed RAPM. You are clueless and you are presenting it very well. You want to talk about specifics? About lineups? Matchups? So do it, because I went over that a couple of times now with either bastillon or drza or whomever. You are late to the party, sorry. And it is quite interesting to see that you are getting so caught up with that, when the numbers are not to your liking, because so far you haven't critized a single post of mine in which I emphasize on the strength of Garnett even though similar numbers are used in order to show that. Why is that?

[mysticbb]

Any statistical analysis that wont accept that premise is even more flawed than the silly per 36 stat that has grown in popularity in recent years.

Yeah, and playoffs are also flawed, because they just don't accept that the best team is determined by the regular season record ...

[MisterWestside]

Seriously, you are confusing things here. I didn't use raw +/-, nor did I use 1yr RAPM. The first number was adjusted for 100 possessions, the next was based on prior informed RAPM. You are clueless and you are presenting it very well.

You're right for once; it wasn't raw +/-. For your second point, you're getting caught up in semantics.

But you're ultimately nitpicking My point still stands. You're quick to explains flaws with data when it doesn't suit your position. Funny how you didn't explain the flaws in your other posts.

And it is quite interesting to see that you are getting so caught up with that, when the numbers are not to your liking, because so far you haven't critized a single post of mine in which I emphasize on the strength of Garnett even though similar numbers are used in order to show that.

Actually, your other posts regarding "the strength of Garnett" didn't simply use one stat (+/-) like you did here.

So do it, because I went over that a couple of times now with either bastillon or drza or whomever.

So why not do it here? Or provide a link? No; you simply put up one number and want posters to blindly accept them. We can read and learn. We know the strengths and flaws of RAPM. And we're not stupid.

[drza]

Thats the thing--you accuse us of ignoring facts but both of us can pick the facts that support our guy. Why is your statistical superiority better than our superiority at the reason they play. Its not fantasy basketball because if it is I concede.

That is a great, Texas Chuck thinks I'm a Garnett/Celtics homer and you think I'm a Nowitzki/Mavericks homer. Well, I probably have a dissociative identity disorder, but maybe you, being such a great psychologist can answer that for me better ...

I feel like I'm in the Twilight Zone. Mysticbb is being called out...for not giving enough support to Dirk??? Dogs and cats, living together...mass hysteria!

TC, are you aware that in the years before Dallas won the title (and before Ahouni went DIRK bonkers) mysticbb was the SINGLE biggest Dirk advocate on these boards. He and I used to have straight up wars about Garnett vs. Dirk, each of us secure in the knowledge that "our guy" was being underrated by the masses. Heck, we still do on occasion. I can honestly say that Mysticbb has done more to change the perception of Dirk on this board in a positive way than any other single poster. I might not agree with him all the time, but I respect the heck out of his positions.

Mysticbb being called out by a Dirk fan for not supporting Dirk enough...who woulda ever thunk it?

[mysticbb]

You're right for once. It wasn't raw +/-. For your second point, you're getting caught up in semantics.

Sorry, but prior informed and 1yr RAPM are two really different things. The used equation is really different. Saying it would be semantics is just plain and simple wrong.

My point still stands however. You're quick to explains flaws with data when it doesn't suit your argument. You didn't point out caveats with the use of any of the rother numbers.

Maybe that had something to do with the way you stated things? Maybe you should ask about issues or why I used a certain number the next time in order to get the "caveats"? Sorry, but you are acting like I never wrote anything about the issues with stats. You will find multiple posts of mine saying that stats are numbers which have to be seen in a context. Maybe it is just a question of time that I don't write a long paragraph about the issues with stats everytime I use a number in one of my posts? Did you ever thought about that? No, obviously not ...


The main point I wanted to make is that players are more than just names, they play differently and by that can impact the game differently. And while Garnett often had not much help on the Timberwolves, we at least can say that the 2005 Timberwolves were actually playing not that bad without him. For whatever reason. In the same sense Nowitzki did not win the title in 2011, the Mavericks did. Just a simple thing. Using the rather simple off number per 100 possessions or the weighted average prior informed RAPM is illustrating something, and I don't see any reason to not use such numbers even with the knowledge that issues are there. And no, that doesn't mean that I think RAPM would be flawless, not at all. It is just simply the best available non-boxscore method we currently have.

Actually, your other posts regarding "the strength of Garnett" didn't simply use one stat (+/-) like you did here.

What? I just used the +14.3 per 100 possessions the Celtics had in 2008 with Garnett on the court in order to demonstrate that the Celtics with Garnett were better than any Mavericks team with Nowitzki. Honestly, you are either a bit selective with your reading or you are just lying to hide your agenda.

[MisterWestside]

Saying it would be semantics is just plain and simple wrong.

Well you didn't exactly help when you didn't state that you were using prior-informed RAPM in your post. Now did you?

I just used the +14.3 per 100 possessions the Celtics had in 2008 with Garnett on the court in order to demonstrate that the Celtics with Garnett were better than any Mavericks team with Nowitzki.

Didn't even see that post. I'm talking about your year-by-year list.

Maybe that had something to do with the way you stated things? Maybe you should ask about issues or why I used a certain number the next time in order to get the "caveats"?

No, you poster the on-off numbers and expected me to say "ZOMG YOU'RE RIGHT GARNETTZ TEAMMATES AS GOOD AS DIRKZ!" without raising questions. Or the prior-informed RAPM (yea, it works to "adjust" for liueups/opponent but it isn't perfect. KG played under two head coaches and often on crappy lineups; RAPM doesn't perfectly account for those things). But look what happened when I randomly gave you the 10-year RAPM numbers about Dirk and Terry and told you to blindly accept that And you took the bait. You KNOW you shouldn't simply give one from of data (+/-) to present your case. It IS useful to our understanding, but there's more to the game than that.

[mysticbb]

Well you didn't exactly help when you didn't state that you were using prior-informed RAPM in your post. Now did you?

No, indeed, I should have said that. So, sorry for not making it clear what kind of number I used.

[kasino]

2000-KG(don't know why this year was left out both were starters)
2001-KG
2002-KG(Nick Van Excel was the difference maker in the best of 5 1st round)
2003-Dirk
2004-KG
2005-Dirk
2006-Dirk
2007-Dirk
2008-KG
2009-Dirk
2010-KG
2011-Dirk
2012-KG

[Texas Chuck]

Any statistical analysis that wont accept that premise is even more flawed than the silly per 36 stat that has grown in popularity in recent years.

Yeah, and playoffs are also flawed, because they just don't accept that the best team is determined by the regular season record ...

honestly have no idea why this is directed at me--Im the one who wants the matters decided on the court in real time not by some statistical simulation telling me which team will win more games if we play the season out X times. I think its terrific the 8th seeded Knicks made it to the finals and the 6th seeded Rockets won a title. I dont want to crown the regular season champ.

[mysticbb]

No, you poster the on-off numbers and expected me to say "ZOMG YOU'RE RIGHT GARNETTZ TEAMMATES AS GOOD AS DIRKZ!" without raising questions.

Honestly, you couldn't even figure out what kind of numbers were used, but you are so sure what my motivation and expectation was? No, my expecation was that you just say: Yes, indeed, we have to look how the players really played and not just go by names. Because that was the point I wanted to make. If I would had any other intention, I would for sure not have answered WhateverBro's question, which he had in the very next post.

You make a simple mistake here, you assume that you know me and my motivation, but matter of fact is that you don't. Then you go on basing your entire argumentation on the false idea that you know exactly what kind of response I expected. You know some posts on the internet, you don't know me good enough in order to make useful assumption about my motivation. The same can be said about Texas Chuck, who assumed based on a couple of posts that I would be just a crazy Garnett/Celtics fan who tries to spin reality.
And without realising it you ended up making posts which rivaled the level of ahonui06. ;)

[mysticbb]

honestly have no idea why this is directed at me--Im the one who wants the matters decided on the court in real time not by some statistical simulation telling me which team will win more games if we play the season out X times.

If that is the case, why did you say that the 2011 Mavericks were better and more impressive than the 2008 Celtics? Did you see them playing a game in real time on the court? Well, no, you didn't, so, how did you try to "prove" your point?

[MisterWestside]

Honestly, you couldn't even figure out what kind of numbers were used, but you are so sure what my motivation and expectation was? No, my expecation was that you just say: Yes, indeed, we have to look how the players really played and not just go by names. Because that was the point I wanted to make.

I was NEVER going by simply the the names, and no; it wasn't just that. You wanted to also make a point about Garnett's teammates in 2005 supposedly being better than Nowitzki's in 2011. Which I disagree with, even though you posted the on-off/RAPM numbers (and by the way, I'm not someone who hates using the data). I watched a good amount of that team's games and still have some of their games on tape: that team experienced turmoil (surly players in contract years, head coaching changes), and lineups/players that Flip Saunders often didn't know how to use properly (and often with Garnett playing alongside them). Nowitzki's teammates in 2011 weren't some legendary bunch in comparison, but at least there was more stability and they weren't coached by an idiot. YES, RAPM looks to account for lineups. But it isn't perfect. Once upon a time, I would be someone who would just look at those on-off/RAPM numbers and agree. But that's not my only source of information, FAR from it.