THE BOOK--Playing The Percentages In Baseball

I know he is “Bill James”, and given that I sparked a big argument with him I’m far from the most unbiased source on the matter, but I have really not been very impressed at all with the quality of the work James has been doing lately.  I think if anyone else were authoring this work the response would be significantly more negative.

I think the world has past him by and it is so tough to do “new” work.  He ignores what is going on elsewhere (as he has admitted) and wastes a ton of effort and time(still uses Excel).  I think he is using the website just to try to get new ideas.  He should just take a few months off, read other information and look for questions others are trying to answer, get an SQL and class and start new.  I think most people at one time admired him and we all appreciated his initial advances, but he reminds me of this old computer programmer at my work who still believes FORTRAN is the greatest computer programming language that exists and won’t move forward.

FORTRAN is for dorks. Everyone knows that programming languages attained perfection with COBOL.

Cut Bill some slack. Yes, his work hasn’t been up to his usual standards since he started working for the Red Sox. If we could catch a glimpse of that work, then I don’t think anyone would be saying that the “world has past him by.” And remember, that you will be an old-man someday, like Bill and the FORTRAN computer programmer. So show a little respect and grattitude for the accomplishments of your predecessors.

Well said terpsfan. Especially, the last couple of sentences. It should not be forgotten where baseball statistics were before this man came along. And, it seems like alot of people on this site get alot of enjoyment out of what it is now.

It’s possible to both respect (or even be in awe of) Bill James as baseball’s Darwin, while at the same time be unexcited about and unimpressed with his most recent work.
-j

I think Black/1’s position here is a fair opinion to have:

“I think if anyone else were authoring this work the response would be significantly more negative. “

Jeff/2’s position about Bill’s stubborness is also a fair opinion to have.

We know the opinions are fair because you could support it with reasonable evidence, without making it personal.

terps/4’s position about deference is also fair:
“So show a little respect and gratitude for the accomplishments of your predecessors. “

***

In summary, Bill’s built up enough currency (with me anyway) that I can be kinder to him than I would with other analysts.  However, at no point can that currency be traded for anything other than civility.  I’m not going to ignore or whitewash or do anything but be truthful.

You do your best to educate, but that only works with an eager student.  I am an eager person on both sides.  Bill is apparently mostly eager on one side. 

In any case, I would never want anyone to cut me any slack.  And, if you take a position against me, trade in the currency I built with you to let me know, so I can have a dialogue about it.  I’ll engage anyone, any time.

With Bill and with BPro, that does not seem to be the case.  Deference can only go so far if they refuse to engage you.

Tango is right when he says the same criticisms we make about Bill can also be applied to the folks over at Baseball Prospectus. You cannot present a closed system anymore and just be done with it. For instance, Bill should of refined win shares by using TT Baseruns for batters, and fixing the problems with how defense and pitching is split. Win Shares terribly under-values great pitchers. I think Pedro was only awarded 29 Win Shares in 2000. Clay Davenport should linearize SB and CS in EqR. It doesn’t make any sense to have a run estimator that is linear for the batting events, but dynamic for SB and CS. Patriot has shown that the run-values for the batting events go down, as you add more SB and CS.

I thought that Bill James was against the principle of an offensive positional adjustment. I remember reading this in Win Shares or the New Historical Abstract.

What was James using Fielding Percentage for? I don’t subscribe to his website. About a month ago, MGL showed that fielders with more range made fewer errors. So using Fielding percentage to approximate defensive value would be similar to using OPS to approximate offensive value. Of course, the defeciencies of using Fielding Percentage would be similar to the deficiencies of using OPS. But both OPS and FLD% can be found just about anywhere, and do a decent job measuring offensive and defensive value.

In the James comments, I alerted him to MGL’s UZR at Fangraphs, which uses BIS data.

So, we give him a pass because he might not know. He knows now. BUT, if he continues to use Fld% and shuns UZR, then I don’t see any reason to show mercy.

"So using Fielding percentage to approximate defensive value would be similar to using OPS to approximate offensive value.”

Probably more like using batting average to approximate offensive value.  There will be a positive correlation, but you’ll be wrong in a lot of cases.  With OPS, you’re still right the vast majority of the time.

I agree with Rally.

Yes, using fielding % to evaluate defense would be more like using batting average to evaluate offense.