Paul Oberjuerge header image 2

Stat King Bill James and Another ‘Baseball Abstract’?

April 22nd, 2017 · 1 Comment · Baseball, France

Let’s get at this the long way ’round.

Most of our possessions arrived in southern France, from southern California, in one big crush of boxes along about mid-October. We had them stacked on a big plastic sheet in the garage, and the boxes made for a pile about three feet high and maybe 10 yards wide and 20 yards deep.

A lot of stuff, and “books” were not a high priority at the moment, considering we did not have any bookshelves.

Not until March did we finally attack the final 15-20 boxes, and I came across my cache of books, a subset of what I had when we left Long Beach for Abu Dhabi, back in 2009. (You can’t travel with a library.)

I was happy to see many of those books, including a half-dozen baseball tomes, led by a book I have cracked open at least 50 times in subsequent weeks:

The New Bill James Historical Baseball Abstract. All 998 pages of it. Still fun as can be. Having it is like rediscovering a long-lost friend.

And as I pick it up and read this section about “baseball in the 1920s” and that section on “ranking the top 100 second basemen” (Joe Morgan, No. 1) … I marvel at how all-encompassing the book is, how amusing, how provocative, how persuasive is Bill James‘s statistics-based writing is …

And then I want a new one. A New New Historical Baseball Abstract, picking up at where he left off, in the one I have in my hands — basically at the end of the 2000 season.

Hey, Bill James! Get busy and update this with another couple of decades! What else do you have to do?

Then I found out … well, the man has lots and lots to do.

A rumpled everyman from Kansas, who pretty much invented the data-driven analysis of baseball, has become a prominent figure among U.S. pop-cultural history cognoscenti, a renaissance man for the computer age.

Since 2003, he has been a “senior advisor, baseball operations” in the Boston Red Sox front office, and the franchise has won three World Series since then — after an 86-year championship drought. A coincidence? I think not.

Many give James much of the credit for the Red Sox signing of free agent David Ortiz in 2003, the addition of a huge talent that changed modern baseball history.

Also, since 2001, James wrote the well-received “true crime” book entitled Popular Crime: Reflections on the Celebration of Violence. dealing with notorious U.S. crimes from Lizzie Borden to O.J. Simpson and JonBenet Ramsey.

And this summer, a book he has worked on for most of this decade investigates a ghastly series of murders more than 100 years ago, which James holds were committed by a single individual — whom he believes he has identified. The book is entitled The Man from the Train: The Solving of a Century Old Serial Killer Mystery.

And in his “spare” time, Bill James works on his website — billjamesonline.com — a sort of open forum in which James waxes prolific on baseball but also on the criminal justice system and other topics that occur to him. Fans can interact with him, and he often addresses specific questions raised. The site also publishes a record of his speaking engagements, many of them made in academic settings.

Getting back to baseball, which is where James made his reputation, I was keen to find out what has happened to the “win shares” system of evaluating the overall value of a ballplayers. James invented the system and used it as the standard method for evaluating players in his Historical Baseball Abstract.

In a podcast made earlier this month, he conceded he had found problems with his “win shares” system use in the 2001 book. But he also, almost casually, took issue with the Wins Above Replacement methodology. WAR (as it is known) has seems to have taken charge of the data-driven overall evaluations, over the past decade, making many converts — while James has been off doing other things.

Toward the end of a 49-minute podcast recorded earlier this month by The Federalist site, James was asked by a caller if he is content with his “win shares” system. In his answer, the Godfather of baseball big data, casually denigrated WAR, which must have set off alarm bells among the ranks of sabermetricians.

He said:

“When I developed win shares I thought, ‘OK, this is the end of that road.’ But then I realized I had made a lot of mistakes in win shares.

“I hope to get a new version of win shares, loss shares, and release it to the public, hopefully this year, but it’s a big project and I don’t know if I will get it done.”

Asked about the basic concept, he added:

“You’re trying to take account of everything a player does on the field that contributes to a win or contributes to a loss.

“The method that is most used now to do that is the WAR system — wins above replacement. But if you look at the two WAR systems that are used, they are actually not that sophisticated.

“I assumed, until looking at it, that these were clever methods that took everything into account, but they’re actually methods that leave a lot of things out. There also are a lot of debatable issues in how to interpret evidence.

“It’s just a question of  working through small issues, one after another, always trying to keep in mind where these small issues fit inside the larger issue of ‘what did this player do to help his team win’?”

So, Bill James may just have declared a stat war on WAR.

But maybe, after that, Bill James can give us a historical abstract that will take us through the 20-teens?

I would love to see one. I hope he gets around to it.

Tags:

1 response so far ↓

  • 1 Doug // May 6, 2017 at 8:36 pm

    I recently read 2017 Bill James Baseball Handbook which is absolutely loaded with stats I didn’t know existed. I prefer the old Bill James Baseball Abstract, but the Handbook is worth a look, particularly if you are crazy about stats and analysis. http://actasports.com/bill-james-handbook-2017/

Leave a Comment