Friday, May 20, 2011

Baseball by the Numbers

Since we're clearly relaxing our way into summer (at least on this blog), I thought I'd talk about sports again. But really, I want to talk about statistics.

I got to thinking about baseball statistics last month after reading a post by one of the internet's brightest---Tim Carmody at Snarkmarket--- about the origins of Rotisserie/Fantasy baseball and the way we read games as culture.

The conversation about fantasy sports often intersects with the story of Sabremetrics---that is, the study of baseball by the numbers, the recent founding of which is traditionally attributed to the baseball writer Bill James and his famed Baseball Abstract.

My hunch is that Sabremetrics first erupted into polite culture by way of Steven Jay Gould. In his famous New York Review of Books essay on Joe DiMaggio's streak he wrote:
Among sabremetricians1—a contentious lot not known for agreement about anything—we find virtual consensus that DiMaggio’s fifty-six–game hitting streak is the greatest accomplishment in the history of baseball, if not all modern sport.
The footnote leads to an explanation of the origins of the term in the acronym for the Society for American Baseball Research. And the next footnote shows that the Daily News had previously introduced the world to Sabremetrics, but with a less august framing. The story was titled: ""Buncha Pointyheads Sittin' Around Talkin' Baseball."

Sabremetrics has now become substantially more well known. Michael Lewis' Moneyball can be held responsible to a large degree. Fantasy sports in general has undoubtedly been an excellent educator. And the 2008 US presidential election included its own advertising for Sabremetrics in the form of the popular, statistical analyses of Nate Silver, a Sabremetrician who turned his skills to politics at fivethirtyeight.com (now a NY Times sub-domain).

It is not surprising that most of the attention to Sabremetrics surrounds the creation of new and interesting metrics or the way able statisticians can interpret and mathematically test vast dumps of data. When we talk about statistics these days, we generally mean a way of thinking or a discipline related to mathematics.

I tend to think about statistics through a mid-nineteenth century lens, however. My statistics happen at a point when some people (ahem, Quetelet) were doing something like what we mean when we talk about "statistical thinking." But most "statisticians" were essentially collectors---like naturalists, but on the look-out for numbers rather than skulls or species. The American Geographical and Statistical Society, founded in 1854 New York as one of the earliest American statistical organizations, neatly folds the statistician in with the explorer. Check out the society's statement of its objects here.

So when I think about the history of baseball statistics, I wonder: where did the numbers come from? Who collected them? Who aggregated them long before Bill James was born?

One might suspect that baseball teams were the first to pay attention to statistics. After all, they were the ones managing their teams. That could well be. But I doubt they are too important to this story.

Individual fans are also likely culprits. But that depends on when it became common for fans to keep their own scorecard. I don't know when that was. I do know that baseball games are *much* more fun if you keep score.

But credit for the popularity of baseball statistics clearly belongs to the press and more particularly to the box score. The New York Times published a brief info-graphic on the evolution of the boxscore a few years ago. And Wikipedia ties early baseball statistics to Henry Chadwick and his publications for the famed dime novel purveyors, Beadle and Adams.

Why and how did the box score become a must for newspapers? I don't know. But I think with that answer and some further digging we can also get some new insights into the way that Americans became more and more invested in a world suffused with numbers and statistics.

6 comments:

  1. The impression I get is that the conventional narrative is that Chadwick basically came down from the mountaintop to the newspapers with tablets containing the traditional stats in the late 19C, and then nothing whatsoever changed until Bill James showed up 100 years later. I think traditionalists like this because it makes them heirs to a long tradition, and sabermetricians (sp., BTW: it's changed since Gould) because it makes them seem more revolutionary. I bet that story is an exaggeration, even if you push it forward a bit to include stuff done by the Elias sports bureau in the teens. If I were going to write a history of the baseball stats, I'd focus on the 50s and 60s, when I suspect a bunch of sentimental white guys started getting all excited about 714 and .366 and 511. (Negro league stats are notoriously bad, which suggests that the demand for reliable batting averages wasn't completely widespread in the 30s and 40s). Strat-O-Matic came out in 1961.

    The other thing that's enormously important for dissemination of stats besides newspapers, of course, is baseball cards: at first, they didn't have numbers on the back, and even in the 50s, cards didn't have the full lifetime stats like the oldest cards that ever made it into my collection as a kid.

    ReplyDelete
  2. I swear I spelled Sabermetrics in the modern form in my first draft, but Gould's essay caused me to switch. I'll amend the entry to reflect the better spelling. Gould's is clearly too pointy-headed and high-falutin'.

    Great point re: baseball cards. I have a Topps Lou Gehrig card printed on a drinking glass (by McDonald's in the 90s) that has his full lifetime stats on the back. But clearly that is an anachronism in all kinds of ways.

    I think score cards are crucial as well. When did mass-produced score cards make it into stadiums, for instance? And what did fans do with score cards after the game?

    ReplyDelete
  3. Interesting column. If you wanted to find out about the early spread of the appeal of numbers, perhaps it's worth checking out the practice of "scoring" the game from the stands. Why was it that people found it worth buying scorecards and then tracking every play in the game? On the surface, it sounds like a pretty odd thing to do. But a good cultural explanation might help illuminate this issue and push it much further back than the rise of fantasy sports.

    ReplyDelete
  4. Yes, Chris, I'm with you. It isn't enough to think about the score cards themselves. We also need to think about what people did with them.

    My hunch---based on no evidence at all---would be that scoring started as an ingenious marketing innovation. I learned to score from the instructions that came with the score card. I suppose we should look at early score cards for evidence. Did scorers need to be taught to track every play?

    ReplyDelete
  5. I should say that I know *absolutely nothing* about baseball (both its history and its current state of play), so what follows may (forgive the pun) come completely out of left field.

    I wonder if there might be a class dimension to this story. I know baseball is the great all-American sport, but was this always the case? Could it have been associated with more moneyed classes early on in it's history? (One reason I think this may be the case is just it's perhaps superficial similarity to cricket which I believe continues to be a very upper class game in the UK.)

    If it was indeed once a more upper class game, and then eventually it became more widely popular, could it be that elaborate schemes for keeping score and calculating statistics emerged as a way to distinguish between *serious* aficionados and the unwashed masses out in the bleachers?

    I guess the argument could work even without relying on economic class. In general, as things become more popular some people look for ways to distinguish themselves. This is certainly something that's constantly happening with popular music.

    ReplyDelete
  6. Hey Lukas. Sounds like an excellent theory to me. I like the analogy to music. Perhaps scoring worked to make baseball more high-brow.

    We have some excellent working hypotheses. Now we'll just have to get some data.

    ReplyDelete