from Amazon
Saturday, March 13, 2010
More Bill James (Spun) Gold
Also in The Gold Mine is an essay on page 49 entitled, “Explaining Defensive Win Shares to a Deceased Sportswriter. I assume the schtick is akin to explaining it as if you were speaking to a 6 year old child - IOW, so that anyone can understand it. After all if an anti-stats (according to Bill) sports journalist from the 50’s can understand it…
Well, that may have been his intention, but I don’t follow his explanation at all. I really don’t. Not at all. Maybe someone can explain it to me.
Bill’s premise is this:
If two fielders are exactly the same on defense on the field (such as if they had the same UZR at the same position), but one is the better hitter (uses up fewer outs on offense, as Bill puts it), then that player has more defensive value. Huh? Keep in mind that I am not saying that it is not true - only that I do not understand his logic.
James says:
.So-and-so is better in the field because he makes fewer outs on offense (even though both players’ defensive stats are exactl;y the same), because making outs increases the team’s responsibility to play defense. Therefore if two players are the same in the field but one of them makes more outs (on offense), the one who makes fewer outs has to come out ahead when you compare the player’s defensive contribution to his defensive responsibility
I think this is only true if you go by “defense per PA on offense” or something like that, rather than comparing players based on defense per innings played or something like that. And I think it is also only true if you consider any defense at all “credit.” If you only consider defense that is above average “credit” then I think that if a bad offensive player is a good defensive player, then he should get more credit on defense because he limits the amount of time he spends on the field per PA, but that is only if you consider a team full of players like that, as in his RC. Or something like that…
Maybe James is just a much deeper thinker than I.
Speaking of James and deep thinking…
For some reason, he continues to slight me. I don’t think he has ever mentioned my name in print or in person (as far as I know). For example, in the same article, he says this:
.(Referring to advanced fielding metrics and the people who “invented” them) Like John Dewan, or Tom Tippett, or Tom Tango, and Dave Pinto
He might as well have said, “but not Mitchel Lichtman.” Now, I don’t care. I really don’t. In fact, I find it amusing. I have actually had several cordial e-mail conversations with Bill. I spoke to him briefly in person a couple of years ago and when I mentioned who I was, he acted as if he had never heard of me.
I don’t know whether any of this is intentional or not (my guess is that Bill would say that it is not), but I think it might have to do with the fact that I have criticized him many times over the years. Nothing harsh, I don’t think, but definitely criticism.
Why have I criticized him? First of all, let me say (and I have also said this many times) that I think that he is a pioneer, a brilliant thinker, and a very good writer, and I owe him an enormous amount of gratitude for inspiring what I love to do - study baseball. Like most saberists, my early inspiration was James, along with Pete Palmer.
That being said, because he thinks outside the box quite a bit and is not afraid to articulate those thoughts, and because he is such a creative thinker and writer, he is going to be subject to criticism. I am just the opposite when it comes to my thinking, my writing, and my work in general. It is cut and dried, I don’t like to posit anything unless I am pretty certain it is true, I don’t like speculation, and when I am not sure of something, I make sure I say so. I do not consider myself much of a creative or deep thinker. In most case, I take other people’s ideas and run with them. That is just the way I am. Because of that, I tend to insulate myself (not intentionally) from substantive criticism. I get plenty of criticism about my style and perceived personality, but those are usually just ad hominem attacks - who cares about those?
I’ll end with one last criticism of Bill’s work so that he can never breathe my name as long as he lives. 
He has an essay at the end of The Gold Mine. I think he printed the same a while ago on his web site. It is called “Cooperstown and the ‘roids.”
If you have not read it, he basically says that 30 or 50 years from now, everyone will be taking steroids to prolong their life and that we will look at steroid users in baseball as pioneers rather than pariahs and they will all be in the HOF (the ones who are worthy of course, independent of steroid use).
I believe his premise in general is preposterous, although it is possible. In particular some of the (speculative) claims he makes in the article are beyond preposterous. I am not really sure why he writes stuff like this. I don’t think it adds to his reputation to be honest. Then again, it is perhaps an example of how he is not afraid to speak his mind in a creative and out of the box fashion and perhaps I am being too harsh on him using my personal standards to judge his work - and that might be unfair.
Anyway, examples of statements I consider “beyond preposterous” from his article:
We will learn to control the health risks of these drugs (steroids or the like), or we will develop alternatives to them. Once that happens, people will start living to age 200 or 300 or 1,000...If you look into the future 40 or 50 years, I think it is quite likely that every citizen will routinely take anti-aging pills every day.
Now, I don’t know exactly how he would define “quite likely” but I assume it means “more than 50% likely” and the word “every” is not ambiguous. I will gladly wager my childrens’ inheritance that in 50 years “not everyone will be taking anti-aging drugs” and I’ll even grant him “95% of the adult population” as the equivalent of “everyone,” just to be nice. And I’ll lay him 2-1. Or 10-1. Or whatever he wants, because the chances of that occurring are very close to zero. Bill says “quite likely” and I say “almost zero chance.”
And, at the end of the day, Mark McGwire is going to be in the HOF, and Roger Clemens, and Rafael Palmeiro, and probablyh even Barry Bonds.
Wow! While I don’t think the chances of all of them eventually getting in are very near zero, I’ll also gladly wager everything I have that not all of them will ever make it into the HOF. And I think he has it backwards. If any of them will get in, it will be Bonds, because he is HOF caliber with or without steroids. Clemens is too. Sosa, McGwire, and Palmeiro - we can easily make an argument that they are not.
Anyway, I don’t know why he writes this hyperbole. I don’t think he can possibly mean what he says. Or maybe he does…
I haven’t read the article so I am probably missing some context. However, the idea that the level of a player’s offensive contribution affects the value of his defensive contribution seems reasonable. Assume, for a moment, that we agree that a player’s value is captured by his impact on expected wins and ignore, say, the changing value of a win as a team moves from 70 to 80 to 90 wins and so on. In this case, the value of a run saved (i.e. the impact on expected wins) should change based on the number of runs a team scores and/or gives up, seemingly with the most value on a team that scores about the same number of runs as it gives up. That is, the team that scores as many as it gives up is on a margin that creates a lot of high leverage situations as they are presumably in a lot of close games.
So, holding all else constant, it seems reasonable that a greater offensive contribution would change the value of runs saved. However, I’m at a loss to explain the direction of the effect in the same way he does. Based on my explanation above, the value of the defensive contribution would depend on where the team he was on was at on the runs scored/runs allowed spectrum. Note that this does not mean that one player is a better defender than another, just that the value to his team of his contributions was higher.
I am not sure what he means by “defensive responsibility.” It seems the proper way to evaluate a player’s ability is through their impact on expected wins. There are at least two ways to do that: 1. context-neutral (UZR etc.); 2. context-dependent (as I’ve written above). Either way, it seems we should be valuing players based on their contribution to expected wins not some “defensive responsibility.” He’s a smart man so he probably means something that makes sense, but I don’t know what it is.