The Devil Is In The Details. Ricky Davis, Too.
Simmons: “I hate adjusted plus-minus. I just think it’s stupid. Is this even a stat you guys are messing around with, do you take it seriously, where do you stand?”
Morey: “Well, I think it’s something to look at. I always think of, you know, all these things are just different views, nothing perfect. There’s never going to be any sort of analysis that will isolate how good a player is.”
The B.S. Report Podcast Bill Simmons – The Sports Guy – Page 2 – ESPN.
I continue to marvel at how people think that just because you use a stat for an analysis that it’s somehow proclaiming that is the end-all, be-all. Rest assured, there are guys who take that approach (Winston), but for the most part, any numbers guy is going to tell you that A. you can’t just look at one measure, and B. you can’t just look at the numbers and make conclusions. What’s interesting is that conversely, most “WATCH THE GAMES!” fools will not tell you the same. They won’t say that “well, you should look at the numbers to see if your perceptions are accurate.”
An example: I tend to be pretty unimpressed with Eric Gordon. I think he’s an undersized dude that launches too many 18 footers, and is basically Ben Gordon before he started thinking he was the best and only offensive option on the floor and started hijacking possessions like they were lost tourists.
But then I look at his numbers. And he doesn’t take a lot of 16-23 footers. He just doesn’t. He’s 20th among shooting guards playing 30+ minutes a game per 40 in attempts from there. And the natural thing would be to say, “Well, Matt, it’s pretty clear you suck at watching basketball.” Which is probably true. But it stuns me that people think they’re so good that they don’t need any other information to inform them. If you’re not a professional scout, it’s pretty likely you’re missing some stuff.
Another example, that presents more of the nebulous nature of this stuff: Tyrus Thomas (who we’ll talk about in a post later). I’m talking to Ziller and defending Taj Gibson, who I think does exactly what Vinny Del Negro wants him to do. If you think that’s the problem, that’s fine, but I can’t fault a coach for playing the guy who does what he’s supposed to versus the guy who randomly goes rogue in search of the weakside block, mopes, and in general is a pain to deal with. Just can’t do it. But Ziller points out that his defensive rating is off the charts every year since he entered the league. In short, the Bulls play better defense when he’s on the floor, and that’s just kind of it. He’s got the blocks, he’s got the steals, which aren’t indicative of good defense, necessarily, but when paired with his defensive rating, it’s really difficult to argue with. He’s a big, aggressive defender that blocks a lot of shots and when he’s on the floor, the defense gives up fewer points. I don’t know how to argue with that, even though I’ve watched Thomas get completely lost on plays, thoroughly give up on possessions, freelance way too much, and of course, he was surrounded by good defensive players on a good defensive team.
So which is it? Is Tyrus Thomas an awesome defensive player that just wasn’t getting playing time? Or a headcase that was covered by the Bulls’ defense and Gibson deserved time? That opinion’s going to fluctuate, from person to person, and it’s likely that Thomas’ performance in Charlotte will end up impacting the answer, when it shouldn’t.
The central point here is something Morey talks about later, which is that the teams at least very rarely use these tools to determine whether a player is good or not. They’re descriptors. And people that know what they’re doing use them that way. For example, saying, Player A’s PER is better than Player B’s PER, so Player A is a better player than Player B is both false and short-sighted. But, if you’re examining what player is the weakest of your starting five, and one player has a significantly sub-15 PER, that’s at least going to point you in the right direction. And that’s if it’s a writer, not a team. A team’s looking at lineups, potential compatability, situational stats, etc.
Long story short, the polarization of this issue muddles the value on both sides. If I tell a stat guy, “But I’ve seen him play! He doesn’t defend well!” they respond with an insinuation that I’m pulling the “WATCH THE GAMES!” card that’s thrown out there with no further thought to the implications. If I use numbers to prove a point, I get drowned with “WATCH THE GAMES!”
Hopefully, the Sloan Conference will help provide context, as Morey states in the podcast. Not that I get to go. Sigh.
Oster-Tags: bill simmons, Daryl Morey, ghostfaceziller, numbersandfigures, stats, statsforthecommonman, Taj Gibson, Tyrus Thomas
