ardee wrote:1. 2005 KG was 98% as good as 2004 KG at everything. Nothing suggests he was having a down year. I hate it when people look at a single miserable stat like On/Off or RAPM and let it contradict every logical thing that observation and common sense dictates.
2. Again, you're only going off On/Off. Provide evidence that Kobe was more of a negative on defense than Harden or Melo and I'll be willing to listen. At the rate this is going we might as well rename the project the top 100 players by RAPM/some form of plus-minus.
Wow. So you're really getting upset. I'm sorry about that.
Try to remember first that it's not the rankings that truly matter. It's the discussion. And while you may be sick of hearing this particular thing discussed, there's plenty of other things being discussed as well. I know I've made a point to do a lot of qualitative analysis, and several other "statguys" have as well.
Second, as you feel like this becomes something like a "Top +/- 100" project that this isn't what the rankings actually reflect at all. That was very clear cut the moment Wilt got placed in the 4th spot. In the grand scheme of how everything's weighted, even if you - unfairly imho - label a certain group as a bunch of automatons blindly using that stat, they aren't dominating the project by any stretch of the imagination. They are perhaps a contingent, and in that sense, they are a contingent bringing one aspect of modern player analysis into the fray and helping get people not used to it, used to it.
Now, in general the +/- stuff is new and controversial. And unfortunately (or fortunately depending on your perspective) that makes it get talked about more. In a conversation with people who really don't know much about basketball, maybe we'd be talking about other things, but in this group there are a lot of things that just don't seem to need to be discussed. It's basically taken as a given here that before people speak up they've at least gone over to basketball-reference and done basic due diligence, and most things you see there people can agree about.
The +/- stuff though is relevant and is not well known, so it gets brought up, and when it disagrees with conventional wisdom we end up having back & forth about the stat itself. I'll fully agree I'm sick of having that conversation. I'd much rather see people who look at a particular number and say "that ain't right" dive in and then explain how the number came to be and why they think it's misleading. I do this as a matter of course, and to be fair, I've seen +/- skeptics do it here and my hat's off to them.
To your points:
1. If we assume KG was basically the same KG from '04 to '05, but +/- says he got a lot less impactful, what does it mean? That's the question to ask, and the answer shouldn't be "then we ignore the +/- data".
It should be pretty uncontroversial to say that when LeBron went from Cleveland to Miami the lift he was giving his teams dropped by a lot. Cleveland went from best record in the league to worst, and in Miami LeBron was not causing a worst-record team to be best. How can that be when LeBron was the same player?
Because the actual value a player contributes depends quite a bit on the moving parts around him and how they fit with him.
In the '04-05 season, the synergy just fell off a cliff. Why? Well it's more complicated than any one thing, but when Sprewell issued his "pay me or trade me" trade demand right before the season began, that was a problem. It's one of those things that's insidious too. A player bitching like that might now keep a team from playing well immediately, but the moment other strife hits, the drop in morale happens like a falling stone rather than a feather.
All this is to say that from my perspective, it doesn't make a lot of sense to look at a team in such a psychological bad place underperforming compared to all norms...and to be dismissive when the +/- data shows something in that direction as well.
I think the box score data and the +/- data combined with what we knew from the prior season really tells us much of the story: Garnett was basically the same player he was before, and he'd go out and rack up numbers accordingly, but when the chemistry went south it affected his ability to impact the game nonetheless.
I'm not going to say that I know all this for an absolute fact, heck I expect it's an oversimplification, but I honestly don't even know how else to summarize things that makes any sense. When the chemistry goes bad, team gets worse, and the star seems less effective according to +/-, why wouldn't we assume the two things are probably related? Why would we even think to look at that as a reason why +/- numbers are just weird?
2. Provide evidence of Kobe being problematic on defense. But, you've already seen articles on this in this project. Stuff talking about Kobe losing his man, gambling when he should be focused on man defense, Kobe coasting. I mean first, if that's not enough for you so be it, but stop acting like nothing else is being provided. I don't understand how people can look at drza or ElGee and not see right from the start that they are providing more non-+/- stuff in their posts than almost anyone else sporting any opinion.
But more pragmatically: If that evidence isn't enough, what are you looking for? Because beyond breaking down video, and hearing people's words, what else is there beyond numbers? And to me the obvious thing to look at is +/- data given that the box score doesn't include the stuff we're talking about. If what you're saying is what you really want is to someone to break down the state-of-the-art game tracking data, then you should be saying that explicitly and doing so politely recognizing that's not something most people are really set up to do. Other than that though, I'm at a loss for what you're asking for and I question whether you even know. And, it's fine to be in a place to say "I'm not convinced, and I don't know what it would take to convince me", but when you're in that situation and you react by attacking the credibility of the people trying to provide you evidence, that's not cool.
