Wade/Tmac/Kobe

[Chronz]

Your last sentence is highly, highly unlikely, but it also doesn't matter who understands the numbers better, my entire point is that the numbers are not particularly meaningful (and even when they're translated to actual in game consequences, large numerical differences amount to very small in-game differences). I have quantified my positions, what I haven't done is the work to take those measures, because doing that work would require much more than a few months time of work (and a large team of assistants to collect said data).

Your post ignores the primary idea of my post though, which is that most of you rarely speak to anything actually basketball related, wherein your position is backed up by a discussion of actual basketball. Instead, there is an opinion on a basketball topic and then a reference to the numbers (we could just as easily replace this with any other sport and the content wouldn't change too much except for the name of the players and the numbers being referenced). As a basketball forum, that leaves a lot to be desired, to say the least. There is very little mention of skillsets and how that affects the game and teammates, the opposition, etc. It's all very formulaic and involves almost no mention of basketball (other than the players being discussed being basketball players).

I've known your pov for months when I begged u to substantiate it only to tell me u didn't have the time to do any actual analysis. U haven't even begun to go beyond ur initially flawed opinions that no one cares about, not anyone here and certainly no one in the actual field. I can 100% assure you, you've given absolutely zero reason for anyone to believe you have the slightest clue of what you're talking about. Psd never really had those types of debate, those who did went into some interesting positions themselves or barely if ever post. Nobody uses stats as a holy grail bud, that debate ended decades ago. You're just not gonna get away with absolutely buffoon takes like ur used to. Im old enough to remember you struggling to grasp the word favorite or at worst not understand how to qualify it beyond ur outdated measures.

Show me an example of said post because u provide neither basketball nor data. Just conjecture. Most of us mix in a bit of it with narrative

[Big Moves03]

And you would rather us say one player is better than the other and here’s nothing to support that.

And I would be fine supporting my position without using any numbers whatsoever, because then there’s nothing that could falsify anything I’m saying. It’d all be 100% opinion. I could say Jason Williams was a better player than Kobe and you could say “I disagree because I think Kobe is better at X skills” and I’d say “well I disagree because I think Jason Williams is better at X skills” and then... nothing. That’d literally be the end of the conversation. Because the only thing you’d have to argue your point is opinions and the opinions of others. Nothing objective (and no, saying Kobe is better at X than Jason without anything objective to back it up does not make it objective).


Also, I think you are genuinely ignorant to a lot of actual stats that can be useful to basketball. But it’s really larger than that. It’s a scientists in essence blanket distrusting all possible stats without knowing the methodology of all of them. That’s a generalization no scientists worth their salt would make. It’s not that you don’t use stats, it’s that you’ve entrenched yourself in a position where you will never accept any statistic as having value. It’s pretty sad.

I haven't entrenched myself in any position. I know the methodology by which these statistics are derived and I've pointed out the massive flaws in them. That's besides the point of my post though. My point is that there is often no discussion of actual basketball. Instead, it's usually you guys citing numbers. We could literally take a kid who has never seen basketball before and tell them to look up numbers on a spread sheet and they would be able to draw very similar conclusions to the things you guys say on here. That says a lot. If you want to use the available stats, that's totally your prerogative, but at least discuss some actual basketball when doing so.

[Big Moves03]

I've known your pov for months when I begged u to substantiate it only to tell me u didn't have the time to do any actual analysis. U haven't even begun to go beyond ur initially flawed opinions that no one cares about, not anyone here and certainly no one in the actual field. I can 100% assure you, you've given absolutely zero reason for anyone to believe you have the slightest clue of what you're talking about. Psd never really had those types of debate, those who did went into some interesting positions themselves or barely if ever post. Nobody uses stats as a holy grail bud, that debate ended decades ago. You're just not gonna get away with absolutely buffoon takes like ur used to. Im old enough to remember you struggling to grasp the word favorite or at worst not understand how to qualify it beyond ur outdated measures.

Show me an example of said post because u provide neither basketball nor data. Just conjecture. Most of us mix in a bit of it with narrative

What I substantiated was how flawed and misleading the present metrics can be and why they should carry very little weight. Sure, I didn't empirically demonstrate that what I was proposing was better, but that's also not my responsibility. We've had this discussion before: If something is broken, it's not up to others to fix it simply because they recognized that it was broken.

I primarily talk about basketball when I post. And for the record, all of this is conjecture, analytics is very, very far from a science and will never come close to being one either. Its very essence is conjecture. We don't have any way to directly assess who is truly better than who. It's not baseless conjecture, but it certainly is conjecture.

The difference is that I at least focus on basketball related topics, whereas many of you typically post numbers and say this guy is better than that guy here are the stats to support that with no real explanation (as far as basketball goes) to explain why. Whether you agree with my takes or not is different, but I am talking about basketball. There are also quite a few on here who agree with my posts (even some who disagree with my conclusions). It's problematic that a large chunk of the posts in here could be replaced by simply referring to other players and statistics from other sports. It strongly suggests there is very little substance being posted and a gross over reliance on the numbers doing the work for them (statistics sure get mentioned a lot in here, especially if they aren't the holy grail).

[valade16]

I haven't entrenched myself in any position. I know the methodology by which these statistics are derived and I've pointed out the massive flaws in them. That's besides the point of my post though. My point is that there is often no discussion of actual basketball. Instead, it's usually you guys citing numbers. We could literally take a kid who has never seen basketball before and tell them to look up numbers on a spread sheet and they would be able to draw very similar conclusions to the things you guys say on here. That says a lot. If you want to use the available stats, that's totally your prerogative, but at least discuss some actual basketball when doing so.

It only says something if you’re of the belief that the stats are massively incorrect.

And no, I don’t think you know the methodology. If you do, please tell me how they calculate Box Creation, RAPM, PIPM, etc.


And again, I’m perfectly willing to discuss basketball, we can even get rid of all objective evidence as you want to do. What actual basketball do you want to discuss first? Let’s go with what makes Dominique Wilkins a better basketball player than LeBron James.

[valade16]

I have also personally never witnessed BigMoves actually talk about basketball. His entire argument for why Kobe is better than LeBron was because a bunch of former players said so. So basically, he thinks stats are so flawed we shouldn’t use them but opinions carry heavy weight. And he claims he’s a scientist.

[bowlofrice]

Kobe

[Big Moves03]

It only says something if you’re of the belief that the stats are massively incorrect.

And no, I don’t think you know the methodology. If you do, please tell me how they calculate Box Creation, RAPM, PIPM, etc.


And again, I’m perfectly willing to discuss basketball, we can even get rid of all objective evidence as you want to do. What actual basketball do you want to discuss first? Let’s go with what makes Dominique Wilkins a better basketball player than LeBron James.

I have explained how doing linear mathematics isn't appropriate in a game like basketball (it basically amounts to using basic arithmetic to try to understand a dynamics systems problem, go google it if you don't know what that is). This approach is wholly unacceptable. It only flies in sports because sports aren't really important and very few who actually know about this stuff care enough to bother with it. You seem to think that some data is better than no data, but that's only true if that data doesn't distort the true outcome and the issue with using the wrong approach is that it almost certainly does in this case. Until a dynamics systems approach is taken, we frankly have no idea one way or the other (and anything to contrary, regardless of whether it turns out to be right or wrong, is simply baseless speculation).

I don't have time to explain all of those metrics, but RAMP basically is adjusted plus/minus using ridge regression (and this technique actually demonstrates that some players adjusted plus/minus changes quite substantially from the more traditional approach, showing that the data can change significantly depending on the technique). The issue is still that we are using linear mathematics (albeit more complex than basic arithmetic) to examine a complex systems problem.

I also don't know why we would talk about basketball reasons that Dominque is better than LBJ. He's not. I can explain why in basketball terms, which is the only terms I think we should be using to assess players. LBJ is a much better ball handler than Dominque and could control the tempo and pace of the game in a way that Dominque never did. When this ability is combed with LBJ's size and athleticism it makes him into a potent playmaker who can get to the basket probably better than anyone in the history of the game, which allows him to warp the floor in ways that Dominque never did. This results in defenses having to expend a lot more resources to stopping LBJ than they did with Dominique. I don't know why you brought up this odd comparison, but I certainly never said Dominque was a better player than LBJ.

At any rate, my entire point with my initial post that you guys responded to is that this is the level that we should be using to assess players. We shouldn't rely on a laundry list of statistics to list off. It's a basketball forum, we should actually talk about basketball.

[Big Moves03]

I have also personally never witnessed BigMoves actually talk about basketball. His entire argument for why Kobe is better than LeBron was because a bunch of former players said so. So basically, he thinks stats are so flawed we shouldn’t use them but opinions carry heavy weight. And he claims he’s a scientist.

You clearly havent read or paid any attention to my posts if you think that that's why I think kobe is better than LBJ. The entire argument is based on basketball and how their different skillsets lead to kobe warping the floor more than LBJ (see that, that's an actual reference to basketball and how their abilities affect the game). I won't re-hash the argument here, but I've posted it many times over. Most of my posts are about basketball and how guys play and how they affect the game. My positions are also falsifiable, it just requires a lot more work to do than the way the game is presently analyzed. There's an argument to be had as to why LBJ is better than Kobe for basketball reasons, which is that LBJ is a better decision maker than kobe was and also is able to control the pace and tempo of the game better than kobe and both of those factors are critical (at least in my book) in how a player impacts the game. My position is simply that warping the floor is more important (again this is falsifiable, but requires a lot of work to examine) and kobe had a greater impact on the game in this sense than LBJ and that's why kobe is better. If the right type of analysis was done and it showed this hypothesis is incorrect or not supported and it was an outcome that we could reliably trust, I would absolutely get behind the data.

Whether you agree with me or not, I am talking basketball here. I'm explicitly referencing these players and their skills/abilities and how that impacts the game. I'm not simply listing off numbers and saying "see, Player A is better than Player B". I'm speaking to what these guys do on the court. That's talking basketball and it's content is not something that could be easily replaced if we switched which sports we were talking about. The content would need to entirely change. That's not the case with a lot of the arguments on here where the premise is basically that one player has better numbers than another player.

[NYKalltheway]

What makes the "numbers" even stupider is the fact that any player would have had different stats playing for another team or under a different system or with different teammates. They take it as gospel when it can be the most abstract thing about the game.

[Big Moves03]

What makes the "numbers" even stupider is the fact that any player would have had different stats playing for another team or under a different system or with different teammates. They take it as gospel when it can be the most abstract thing about the game.

Yep, this 1000%. A player in one system might put up completely different numbers in a different system or role. That's a critical factor to consider and something that typically gets completely ignored on here.

[valade16]

What makes the "numbers" even stupider is the fact that any player would have had different stats playing for another team or under a different system or with different teammates. They take it as gospel when it can be the most abstract thing about the game.

Yep, this 1000%. A player in one system might put up completely different numbers in a different system or role. That's a critical factor to consider and something that typically gets completely ignored on here.

See? This is what I’m talking about. You guys are criticizing stats when you know nothing about them. You guys both realize they actually have stats that take into account their statistical fluctuations with different teammates, lineups, etc.

[valade16]

You clearly havent read or paid any attention to my posts if you think that that's why I think kobe is better than LBJ. The entire argument is based on basketball and how their different skillsets lead to kobe warping the floor more than LBJ (see that, that's an actual reference to basketball and how their abilities affect the game). I won't re-hash the argument here, but I've posted it many times over. Most of my posts are about basketball and how guys play and how they affect the game. My positions are also falsifiable, it just requires a lot more work to do than the way the game is presently analyzed. There's an argument to be had as to why LBJ is better than Kobe for basketball reasons, which is that LBJ is a better decision maker than kobe was and also is able to control the pace and tempo of the game better than kobe and both of those factors are critical (at least in my book) in how a player impacts the game. My position is simply that warping the floor is more important (again this is falsifiable, but requires a lot of work to examine) and kobe had a greater impact on the game in this sense than LBJ and that's why kobe is better. If the right type of analysis was done and it showed this hypothesis is incorrect or not supported and it was an outcome that we could reliably trust, I would absolutely get behind the data.

Whether you agree with me or not, I am talking basketball here. I'm explicitly referencing these players and their skills/abilities and how that impacts the game. I'm not simply listing off numbers and saying "see, Player A is better than Player B". I'm speaking to what these guys do on the court. That's talking basketball and it's content is not something that could be easily replaced if we switched which sports we were talking about. The content would need to entirely change. That's not the case with a lot of the arguments on here where the premise is basically that one player has better numbers than another player.

Actually LeBron warps the floor more than Kobe.

Now how does the discussion proceed? I think LeBron’s basketball skills worp the floor more than Kobe’s, you think the opposite. We cannot use anything objective to measure it. So how do we proceed? Is it just whoever yells their opinion is right louder or?

[valade16]

I have explained how doing linear mathematics isn't appropriate in a game like basketball (it basically amounts to using basic arithmetic to try to understand a dynamics systems problem, go google it if you don't know what that is). This approach is wholly unacceptable. It only flies in sports because sports aren't really important and very few who actually know about this stuff care enough to bother with it. You seem to think that some data is better than no data, but that's only true if that data doesn't distort the true outcome and the issue with using the wrong approach is that it almost certainly does in this case. Until a dynamics systems approach is taken, we frankly have no idea one way or the other (and anything to contrary, regardless of whether it turns out to be right or wrong, is simply baseless speculation).

I don't have time to explain all of those metrics, but RAMP basically is adjusted plus/minus using ridge regression (and this technique actually demonstrates that some players adjusted plus/minus changes quite substantially from the more traditional approach, showing that the data can change significantly depending on the technique). The issue is still that we are using linear mathematics (albeit more complex than basic arithmetic) to examine a complex systems problem.

I also don't know why we would talk about basketball reasons that Dominque is better than LBJ. He's not. I can explain why in basketball terms, which is the only terms I think we should be using to assess players. LBJ is a much better ball handler than Dominque and could control the tempo and pace of the game in a way that Dominque never did. When this ability is combed with LBJ's size and athleticism it makes him into a potent playmaker who can get to the basket probably better than anyone in the history of the game, which allows him to warp the floor in ways that Dominque never did. This results in defenses having to expend a lot more resources to stopping LBJ than they did with Dominique. I don't know why you brought up this odd comparison, but I certainly never said Dominque was a better player than LBJ.

At any rate, my entire point with my initial post that you guys responded to is that this is the level that we should be using to assess players. We shouldn't rely on a laundry list of statistics to list off. It's a basketball forum, we should actually talk about basketball.

The reason I brought it up is because NYKalltheway thinks Nique is better than Bron. So if analyzing actual basketball skills is so much better than using stats, how can you explain how NYKalltheway can analyze actual basketball skills and come to a conclusion that Nique is better than LeBron?

So here’s the problem: your basketball analysis technique led to a far more stupid conclusion (Nique > Bron) than the statistical technique ever did.

[Big Moves03]

Actually LeBron warps the floor more than Kobe.

Now how does the discussion proceed? I think LeBron’s basketball skills worp the floor more than Kobe’s, you think the opposite. We cannot use anything objective to measure it. So how do we proceed? Is it just whoever yells their opinion is right louder or?

We could look at stats with different lineups, sure...we can't look at stats in terms of how one player would perform in a situation that they were never in (because we have no idea of how that would really turn out). For instance, we can't scale Curry's performance to a team where he's playing with a dominant big man in a slow half-court offense, because he's never played in that kind of offense.

[valade16]

We could look at stats with different lineups, sure...we can't look at stats in terms of how one player would perform in a situation that they were never in (because we have no idea of how that would really turn out). For instance, we can't scale Curry's performance to a team where he's playing with a dominant big man in a slow half-court offense, because he's never played in that kind of offense.

Sure but you’re saying the problem with stats is when analyzing things that haven’t happened they suffer from the same problem as talking basketball skills: conjecture.