THE BOOK--Playing The Percentages In Baseball

I thought it sucked..really bad.

Of course, I’m coming from the perspective of one who is very familiar with James.

I thought it was ok.

The interview was more geared to the general population so he didnt really say a whole lot that we didnt already know.....wins are bad way of evaluating pitchers etc etc.

My thoughts echo John’s.... it was more an introduction to Bill, with nothing new to expound on.

The funny thing was I was watching the Braves-Nats game, and Jon Miller happened to mention Bill James right as Morely Safer introduced the segment here in the Mountain time zone.

Yeah I heard that as well during the broadcast......something like bill james gave them a list of names who never won rookie of the year yet made the hall of fame....i think they were talking about ryan zimmeran at the time.

I enjoyed it, it was a good piece.

It’s not the biting 60 Minutes investigation that we grew up with, but it wasn’t intended to be that either.

They *could* have had some contrarians that you could find in the mainstream media.  The closest they came was David Ortiz’s advocates.  Way too much Bob Costas that’s for sure.

It confused me when they said Bill debunked the myth that you run on the pitcher, proving that it’s the catcher who prevents stolen bases.  Is that true?  If anything, it’s the other way around, isn’t it?

I am not sure how having the contributions of sabermetrics to baseball portrayed in a positive way on the most respected national news program could ever be seen as bad.  60 Minutes had obviously settled on a point of view that James was the genius behind the Sox’s two world series wins and had edited the piece accordingly.  But with careful wording I think they avoided any outright misstatements even if they left the impression that James was the originator of DIPS theory.  I think that the connection of James’s hiring with the Sox series wins was actually a test for would be sabermetricians on whether they know that correlation does equal causation.

In the 1982 Abstract James discussed steals.  First, looking back, it is amazing how much more data is available now than then.  He was limited to going through box scores and was only able to count steals in games started by a particular pitcher or catcher because there was no way, from the box score, to see if that pitcher had been removed before the particular steal.  Also, CS was not listed in the basic box score, so he had only half the data even in the limited form noted.

(As an aside, his work getting Project Scoresheet [which eventually morphed into both Stats Inc. and Retrosheet] off the ground is, in no small part responsible for the current wealth of data available.)

So, I wouldn’t put much weight at all on the study.

But for what its worth, his conclusion was “They may steal on the pitcher, but they still steal a hell of a lot more against Joe Nolan than they used to against Johnny Bench.  It is true, however, that there are wider differences in the OSB rates of various pitchers than of various catchers, which would suggest that the pitcher can do more to prevent or allow a stolen base than the catcher can.” 1982 Abstract, pg. 13

#8/dlf: Thanks!  That’s how I remember it, that stealing was perceived to be influenced by the catcher’s arm, and Bill showed that the conventional wisdom was not the whole story.

But was I misinterpreting?  Was conventional wisdom really the other way around, that people really thought you steal on the pitcher?  If so, when did that change?

Yes, it must have irked Voros a bit that James would quote Voros’ tenet, making it seems as if James was the originator.  James was sorta the originator, since James popularized DER (not sure if James or Erik Walker was the originator, but both were close).  But James by his own admission had never thought of the 3 true outcomes the way Voros did.  And, I also hear that Ron Shandler may have a claim to fame to it as well.

I think James’ writing, while its magic was not explicitly mentioned, was implied in a few of the conversations.

***

As for the C/P thing, using WOWY:
standard deviation of SB+CS run prevention among the top 170 catchers in games played (average of 30,000 sample PA) was 4 runs per 5000 PA.  Among the top 800 pitchers (sample of 6,400 PA), the SD was 6.5 runs per 5000 PA.

If they were both equal, you’d expect the observed spread to be larger for pitchers, because of the much smaller sample size.  If you account for that, the true talent is probably a bit larger for catchers.  I’d have to work it out.

However, pitchers are NOT selected based on how well they hold runners, while catchers certainly are.  If catchers were not selected based on their arm, the spread would be far wider for catchers.

Catchers are the determining factor, but once you have a league of all great throwing catchers, they no longer become the determining factor.  If that makes sense to you.  The determining factor is based on the spread in talent, not the mean.

It was OK—as others said, clearly geared to the general public rather than, say, the people who post here. I had never actually heard him speak before, so that was interesting.

I was also disappointed that the quality of his writing was not discussed more.

I do recall that Bill James gave Voros McCracken a major hat tip in one of his books on the DIPS issue.

Yes, in his book he gave MAJOR props to Voros.  This 60 Minutes piece made it seem as if James originated it.  It’s all in the editing (which James can’t control), it appears.

***

Feedback from SOSH:
http://sonsofsamhorn.net/index.php?showtopic=29550&st=20

Eric Van has a comment at the end of that page:

***

Here’s a much better inteview about James on PBS:
http://www.pbs.org/thinktank/transcript1197.html

(Hat tip: SOSH.)

But was I misinterpreting?  Was conventional wisdom really the other way around, that people really thought you steal on the pitcher?  If so, when did that change?

This goes back thirty years, but I seem to recall that in the Johnny Bench era, the “common wisdom” was that the catcher controlled the running game, but that Joe Garigiola frequently defended Ted Simmons, whose arm had a horrible - and perhaps undeserved - reputation by assigning the blame to the pitcher.  From the dusty recesses of my memory, I don’t think that this position was the majority position, but with Joe G’s prominence on GOTW back before the superstations and ESPN, it was a loudly stated one from a perceived expert.

I had low expectations for it, and given that mindset, I thought it went pretty well.

I think when you’re familiar with a subject and see it portrayed in mainstream media, it’s inevitable that you’re going to be annoyed by the incorrect details and oversimplifications.

In this case, I think the producers felt obligated to give a nod to the “other side of the controversy” .. so we got to hear about how his ideas about clutch hitting had turned out not to be correct (I’m paraphrasing).  And there was the usual nonsense about “what about the human element that numbers can’t capture?” ... on the other hand, the tone was mostly positive, and you could hardly expect better treatment in the MSM.  As others have mentioned, the piece ultimately went too far in crediting the Red Sox success to James anyways.

FYI, I do think they got the K:BB ratio stuff correct.  While Voros had a nice way of articulating the issue, James is the first guy I can recall emphasizing the importance of strikeouts and walks for predicting future performance .. and that’s basically what was discussed in the piece.

The important thing to remember about the popular press is that they don’t have the time to cultivate any sort of expertise in a subject matter. Tomorrow, Bill James will still be studying baseball; Morley Safer won’t.

The function of the popular press was generally to provide two things - a communications channel, and people capable of communicating well - as an intermediary between experts and the general public. The Internet has provided everyone a communications channel (if not always the ability to write or communicate clearly), and so we’re starting to see a lot more of just how little our journalists actually know about anything other than how to practice journalism.

And, as has been pointed out before, the target for this piece is a general audience - one that probably isn’t particularly passionate about baseball, even.

Haven’t decided whether or not I’ll watch yet.

I updated the link at the top to get you the full transcript.  Those not wanting to watch 15 minutes of it, and wanting to quickly skip over Bob Costas will prefer this.

On BTF, the James thread has evolved into the eternal clutch thread.

There is no question that this statement is true:
CLUTCH SKILL EXISTS

The question on the table is:
Can you identify this clutch skill?  And, to the extent that you can identify it, how much of it can you identify, and how much time do you need to identify it?

IIRC, you need something like 5000 career PA to be able to get an r=.50.  That is, the following two statements gives you the same “certain amount”:

1. after 200 MLB PA, I can identify a certain amount of overall batting skill for a player

2. after 5000 MLB PA, I can identify a certain amount of clutch batting skill for a player

Is any of this in dispute?  If so, then change “5000” in number 2 to whatever number you want.  I’d like to know what you think it is.

For those who disliked it, did you honestly expect James to get on “60 Minutes” and discuss regression formulas, or any type of high-powered mathematical equations?

If you want more people to become interested in “our way of thought” you need to ease them in, not overwhelm them from the start.  Hopefully people watched this and, if they did not previously realize, begun to understand simple ideas like the W-L record is bad for evaluating pitchers and such.

We can talk about how we dislike that it did not tailor towards the sabermetrics community but I HIGHLY doubt “60 Minutes” decided to go with a James segment for “us.”

Those that think along the lines that we do come to blogs like this and talk; those that do not are generally the audiences that a show like “60 Minutes” caters to.  It introduced some sabermetrics topics to an audience that likely had not heard much about it or been extremely intimidated by it.

Was it perfect?  No.  Was it a step in the right direction?  Definitely.  If we’re going to get angry when commentators and fans overemphasize the wrong stats we should be doing what we can to break them in, not critiquing the attempts to help.

Yup, I thought it was pretty good, all things considered.  James has a certain charm when he speaks and generally knows the “right” things to say in the various forums and contexts.

Costas, while definitely a smart guy and one of the more knowledgeable MS sports people about lots of things, including some sabermetric principles, comes off like a pompous ass.

I couldn’t believe that he actually said something like (I am REALLY paraphrasing), “James merely confirmed or added to the breadth of what I already knew or suspected.” Funny, I thought the same thing when I learned in college about Einstein and his theories on special and general relativity.

The thing about clutch and any other “skill” in baseball, the proper question is not, “Does it exist?” It is, what is the spread of the skill among the population, or some such thing.

Half (or more - actually maybe 90%) of the controversy in the world is a lack of understanding of the material by one or more parties, or a lack of relevant knowledge by one or more parties - NOT a genuine disagreement about philosophy.

I mean, seriously, what is it about clutch skill that we DON’T know?  Nothing, as far as I am aware.  Not “finding” it in prior research (say, prior to Andy’s in The Book) is not a mistake as in, “We screwed something up.” It was a natural mistake that happens when something is small and you are trying to find it in limited sample data.  And not a large mistake at that.  People don’t realize that the gap between saying, “It does not exist,” and, “It does exists, to the extent that it does,” is so small as to be almost meaningless.  If someone wants to get excited over the fact that researchers said that it did not exist and now they say that it does, that’s fine by me.

That is A LOT different from the sac bunt stuff.  When sabermetricians said that the sac bunt was almost always a poor strategy, they screwed up royally, by doing a sloppy job in their methodology, and/or by not specifying what they meant by “a sac bunt.” Then again, even that was not a catastrophe, as at least they either said, or it was obvious, that their conclusions were only based on, though it should have been made crystal clear, a successful sacrifice bunt where an out was made and a runner was advanced.

I agree with mgl on clutch.

As far as E Seidman’s comment that the James segment is “a step in the right direction"--I don’t really care at all if sabermetrics becomes ‘popular’, or generally accepted, or not. Yes, it can be irritating to see managers and players and sportswriters saying silly things. But OTOH it’s kind of fun to be ahead of the curve, and to have an ‘advantage’ on the general public. All I care about is that there is enough of a demand that the supply of sites like this, and books, and serious reseachers is enough to keep me ‘entertained’.

The reason I didn’t like the James interview is not that it didn’t go into ‘baseball data analysis’ enough--I didn’t expect it to. It’s that it didn’t go into Bill James enough. I learned nothing about him that I didn’t already know. And a casual, unfamiliar viewer would have little reason to seek out James’ stuff on the basis of that weak interview.

Again, on the mgl clutch comments--I think his point is important because of its practical nature. I mean, if you are obtaining a player who has enough PAs to be confident that he can hit a bit better in the clutch, then he is already in decline, and you better pay more attention to that. And if you are looking to increase your enjoyment watching a game, the apparent clutch impact is so small that the impact on a random clutch PA situation is completely negligable.

So, the only relevance that I can see is in a HOF type evaluation.

For me it just really depends on what I consider sabermetrics and statistical analysis to be.  I don’t conduct research or spend hours upon end for no money looking at game logs from 1973 to be different or come up with cool and new metrics.  Rather, I do it because I’m hoping that my work, in conjunction with that of others, can help shift the thought process in baseball “strategery.” As in, I, as well as many others search for the truth and write with the greater good in the back of our minds.

I feel like if we possess the mindset that we do this just to be ahead of the curve we’re falling more in line with punk music - a style that stakes its major claim as being different than the norm.  I just really feel that sabermetrics should be along the lines of what we find in “The Book” - a search for the truth and analysis of strategies in order to help the greater good of the game, not just to be ahead of everyone else.

I have a friend who knows nothing about sabermetrics except on the rare occasion that I explain a particular concept to him, and he felt that the piece was incredibly vague and that he learned nothing from it.

Eric has a point that anything technical really had to be excluded from it to cater to the masses, but I don’t think that anyone really took anything away from the piece.

You know, maybe David is right.

Just now, on the Science channel, was a program called “Baseball’s Secret Formula”, and it was one hour on Bill James and Sabermetrics.  This show was far better, and James was all over it.  And instead of Bob Costas (yech), there was Alan Schwarz.  Much better choice.

It’s repeating on my cable network (Cablevision) three times on Tuesday morning.  I think it was done in 2005 or 2006.

***

And, cool to me, while there was no citation of it (I think they made it seem as if James produced those numbers), they actually referenced this chart:
http://tangotiger.net/RE9902.html

Only I would know really, because I use those numbers so often.  Print out that chart, and when you see them talking about it, you’ll get a little chuckle.

I’m going to take a little contrarian viewpoint here. First, having 60 Minutes do something, anything on baseball that wasn’t steroid related was fabulous. Mainstream media, top ten show, etc., even if their average viewer is 194 years old.

What the piece did was present this average guy who ended up with one version of the American dream--changing baseball and working for an MLB team without being a stud player. The details and the stats weren’t the story.

It was good for baseball, which is good for all of us because it puts money in the coffers. It was good for James and SABR because it showed a regular guy, with a quiet sense of humor, instead of the stereotypical nerd in the basement. Costas may annoy the hell out of us, but he’s America’s sports guy and his presence lent credibility to the whole numbers thing.

I agree with everyone who said it’s incredibly frustrating to watch mass media stories about things you know well. It’s pretty typical to not really get details wrong, but kinds miss the point. Having said that, for the audience they have, I think the story they did was dead on.

I’m not sure what’s so contrarian about that.  Post 25 seems to be the majority opinion.

It was late when I read through the comments, and the points against the broadcast stuck in my mind. Never post when you’re too tired to have a decent conversation.

http://cdn3.libsyn.com/tsoya/tsoya080429_billjames.mp3

Excellent interview of Bill James.  Near the end, when asked to name the current crop of analysts who grew up on Bill James, he cited Hardball Times!  And specifically noted Greg Rybarcyzk’s shift article (though not by name).  He also said he couldn’t do what Greg did, that it was beyond him.

And he’s also thinking about the pitchf/x stuff, and trying to figure out how to make sense of it.  He’s probably like alot of people, that is simply overwhelmed with it, and doesn’t know really where to begin with it.

Oh, and he says he never argues, that he just puts out his argument, and either you buy it or not.

Sounds too guru-ish for me, but you’ve gotta respect him for his position.