In The Man from the Train: The Solving of a Century-Old Serial Killer Mystery (Scribner, Sept.), legendary baseball statistician Bill James and his daughter, Rachel McCarthy James, use statistical analysis to shed light on an unsolved series of murders in 1912 Iowa.

What is the relationship between your work in professional baseball, specifically the development of sabermetrics, and your approach to history?

I’m the same person writing about one thing as I am writing about another—an analytical person, as opposed to an intuitive person or an emotional person. An analytical person instinctively asks, “What are the elements of this problem? What’s the relationship of one of those elements to another? Where have we seen other problems that are like this one?” That’s just the way I think about anything in my life.

Tell us how the investigation for this book began.

I was interested in the cases that surround the 1912 Villisca murders. Working on them, it was natural to say, okay, let’s walk in this direction and see if we can find the source of this nightmare, just as you might work backward on any problem to try to figure out where it came from. It happened that by walking in that direction, we stumbled across a culprit who sure looks like he’s got to be the guy behind these murders. But there was a lot of luck in that.

Could you apply this method to other unsolved crimes, such as the Black Dahlia murder or the 1885 servant-girl murders in Austin, Tex.?

I think that probably the Black Dahlia was murdered by a serial murderer, and, if that is true, then you probably could identify other crimes that he had committed and walk them backward toward their origin, I would guess. But our guy, the murderer we were working to identify, left evidence of his identity all over his first crime. In a basketball analogy, we threw up a three-pointer off the backboard and it went in.

What limits are there to this kind of research?

Well, this specific approach is mostly limited by the pretty short list of cases for which it would be appropriate. I anticipate that, over time, more and more old newspapers will become available, and that research of this general nature might become much easier. But it might not happen that way. Instead, it might happen that, for one reason or another, the spread of information is arrested, perhaps by investors who take control of the market and won’t grant easy access to it.