Bill James published 40 issues of The Baseball Analyst in the 1980s. All are available on SABR’s website.
Bill James’ introductory note takes delight in the fact that two of these articles are followups on efforts published in earlier editions. Sabermetrics was a new field, back then, and the practitioners needed to cross-pollinate; Bill’s Analyst was a way to make that happen.
Jim Reuter led off with a followup to Jim Morrow’s piece in the second issue; Reuter’s piece was called More on the “True” Slugging Percentage. Where Morrow’s version was a linear weights approach, Reuter began with Runs Created. Predictably, his results resembled RC, as Reuter concludes. The mystery to me is why either writer thought improving SLG was a necessary effort; it’s OK as a rough measure, and if you’re going to recreate RC you might as well not call it Slugging Average. That said, this is a nice piece.
Ward Larkin’s piece is about normalizing batting averages across eras. Larkin uses Standard Deviations and makes a plate appearance adjustment. The results are unsurprising if you’ve read any of the subsequent sabermetric literature, but it was probably worth doing. There’s a good writeup of his methods, but I see no calculations; there is a long results table. His followup comments are excellent.
Dick O’Brien followed up on Dallas Adams’ article about overworking rookie pitchers, and discovered that the effect was probably more complex than Adams’ analysis. He concludes that young rookie pitchers with high workloads are usually successful; it’s pitchers who come up around age 25 and shoulder heavy workloads who develop problems. O’Brien uses a different data set from Adams, which may be pertinent. This article has a couple excellent tables and a handful of paragraphs of discussion; more work is clearly indicated.
Finally, Craig Wright looked at player development, which he defined a bit differently than I would, and concluded that the Scouting Bureau wasn’t harmful, but wasn’t pulling its weight. He also discovered, unsurprisingly, that teams that skimped on player development at the Class A level had poor results. He used James’ Value Added Method as a yardstick for this analysis. I presume the work Craig did for the Rangers often read like this essay. It’s certainly different from any baseball analysis anyone else was publishing at the time.
James added a note at the end, contesting one of Ward Larkin’s conclusions about the shape of talent on major league teams. The interesting thing, to my view, is that Larkin’s charts don’t agree with James’ claims; I’m hoping there was further discussion.
Again: All 40 Baseball Analyst issues are available on the SABR website.
My girlfriend squealed with delight yesterday when telling me about your post. She’s a librarian and has special reverence for the printed word. She’s also a blogger and holds blogging in even higher esteem.
Naturally I had to reread the article. I can’t compare it to current sabermetric research. I gave up on baseball in the late 1980s. In fact, a few months ago someone mentioned Josh Hamilton, and my honest response was “Who’s he?”
Thank you for saying that the followup comments were excellent.
Thanks, Ward; glad you (and your girlfriend) enjoyed the review.
Lots of baseball writers doing long-term analysis are using methods similar to what you describe, these days, and (of course) I’d assume anyone reading my reviews of the older sabermetric literature would have some familiarity with them. Chris Jaffe’s book about manager careers comes immediately to mind; I reviewed it here.
Thanks again for the note.