Doctor MJ wrote:What you're talking about here is a limitation on the precision of the tool, not a fundamental bias. I'll readily concede the former, but it rapidly becomes a smaller and smaller issue when you has much data as we have. Not that you can ever say "His +/- stats say X, he is clearly X good at playing basketball", it's not so straight forward, but if we're talking about issues relating to team dynamics, etc, then we are talking about issues that all stats would have because these are things that actually determine how effective a player is.
What +/- is telling us right now is that Durant has taken a leap forward in terms of how effective he is being. One can argue this has nothing to do with Durant improving (though I know you aren't doing that either), but there is enough data that it can't simply be dismissed for small sample size.
I'll agree with your post here, in principle. You're right; I'm not claiming that Durant's metrics are merely caused by stat social noise.
Sure it depends on your roster, but it's pretty hard to imagine the team whose offense wouldn't get considerably better if it had Durant in place of its lead off-ball scorer. He's better at it than everyone else. Put him on a weak team, it's a huge help. Put him on a strong team, he'll still lift the ceiling. There's basically no limit to how good a Durant-led offense can be.
LeBron on the other hand his offensive game is so impressive because he can do everything, which is at its most astonishing when you have other guys with a clear lack of skills. It's easy to imagine scenarios where diminishing returns are apparent because of this, along with the fact that we've actually seen it in Miami. When you add in that quite frankly there are players i"d rather have on ball than LeBron because they are better playmakers, that takes the portability issue further.
However, I disagree with you here. I don't think you're insinuating that LeBron isn't as "portable" but just in case your are, keep these two things in mind:
1. LeBron has played more off-ball basketball his season than in any other season, and production-wise he's in the neighborhood of his ridiculous '09-'10 campaigns. He's is more than capable at shifting his game to adapt to his team.
2. LeBron plays on a team with prime/elite NBA talent, and one of those players shares a similar skillset. He also joined a new team os opposed to having the team built around him. And yet the Heat clearly miss LeBron when he isn't on the floor, both on offense and defense (both from the game tape and the metrics). If that doesn't speak to his portability, I don't know what does.
3. In '12, savvy statisticians have observed that even with Durant's superior off-ball skills, his team offense was not as potent when Westbrook wasn't as assertive and Durant had to assume more offensive responsibilities. Even with great off-ball players, you have to assmble your team properly in order to maximize the talent. Being off-ball isn't necessarily "better" than being on-ball; you have to consider the rest of the roster's needs.