Now that I'm back to Chicago and properly freezing, and as I gear up for a quick rat-a-tat through winter moves that have been made—tomorrow, I'll be writing about the Werth and Crawford contracts at some length—I wanted to step away from the Hot Stove stuff for a second and dig into two things that were sort of interesting, not just because they came up, but because of what we might learn in each instance going forward.

  • First, the subject of defense, which I know everybody likes talking about, and which sometimes leads to useful info and insight. I was struck by two separate conversations I had, one with a statistician, and another with an assistant general manager. Between the two, the subject of defensive metrics came up, and both were fairly dismissive of UZR. Maybe that's a matter of reaction formation, of wanting to take the current pop hit down a peg; that goes with the territory of invention.

    Nevertheless, the AGM's remark, in response to my question about who's using what, was the one I found telling: "Nobody uses it—why would they? They've all built their own, better metrics."

    Now, to be sure, the unspoken amendment to that I'd insert was "among those worrying about defensive metrics, or who find value in them." But that said, I started my own informal survey of front-office folks in the spring, asking whether they were using off-the-rack defensive stats, tailoring that kind of data to suit their own purposes, or going whole-cloth in the creation of their own metrics. Nobody would admit to going with off-the-rack choices, and only one said they were tailoring publicly available info; everybody was doing their own thing. And no, they wouldn't share.

    Some teams have been more frank and public about such things—the Red Sox are a good example. However, the more interesting dilemma to my way of thinking is that this reflects a growing problem with sabermetrics itself.  It used to be fairly transparent, either just in terms of output (the point we used to stop at), or as far as both the product and the process of generating it (or what we might call the old, "open source" standard). But the more complicated the tool, and the more pointed its purpose—to achieve competitive advantage, instead of just the dissemination of knowledge—the less likely it will be to make its way into any other team's arsenal, let alone into yours or mine.

    In some ways, this is troubling, because I think all of us have that "kid on Christmas" instinct, to get the answer and take comfort in its surety. But it's also an echo of what has already happened in professional basketball. When I was in sports publishing in the early Aughties, I was happily releasing the original works of Dean Oliver and John Hollinger. What's happened since is telling, and was reflected in the proceedings of the 2009 MIT Sloan Sports Analytics Conference: statheads like Oliver and Hollinger were on the stage, but Oliver had long since hooked up with the Nuggets as their stathead. They joined the Mavericks' Mark Cuban, the Celtics' house stathead, and more. Hollinger was the one person onstage in the fourth estate, while everyone else was working for a team (or owning one).

    The takeaway, both in terms of who was on stage as well as the fact that all of them (save John) couldn't really get into their own methodologies for basketball performance analysis, was that the sabermetrics revolution in basketball has happened, but was bought up almost instantly. That's because the products of performance analysis were welcomed by industry, who promptly bought as much of it as they could, long before it achieved anything like baseball's popularity. It's sort of like the Founding Fathers chucking rebellion when they could instead get bankrolled as consultants to the Crown on the benefits of taxation with representation—it's a reasonable choice by individuals, but one that leaves the entertainment/consumer section of the market less well served.

    Happily, this is where Kevin Pelton and company ride in on that side of things, but the fact that this is becoming more and more of a factor in baseball research, where the researchers themselves and the best work never reach broader dissemination, isn't really a happy outcome, is it?

  • The other thing I wound up enjoying a lot of was talking with women in the industry, running into them, or just seeing how many were working—the halls, the media room, down in the vendors' showroom, you name it. Compelling examples like seeing Justine Siegal and her crew from Baseball For All running their first-ever booth on the showroom floor, or bumping into Kim Ng of the Dodgers, were easy, but catching up with Lisa Winston is always fun, and finally meeting Brittany Ghiroli—one of last year's BP Idol contestants—in person was nice as well.

    But a compelling reminder of how far things have come came from a conversation with former Cubs front-office worker who talked about former media director Sharon Pannozzo's lot when she originally joined the organization in 1982/83. His comment was that she was just the second woman hired by a front-office (in a non-clerical job), and the early years weren't easy—he mentioned Lee Elia's spiteful little choices, like making her take a cab instead of traveling on the team bus when they were on the road, since he didn't think she belonged with the ballplayers like the other front-office employees. That was less than 30 years ago. This was interesting to me because a lot of people in the media—usually anonymously, but in Terry Boers' case, not so much—liked to complain about Pannozzo in later years; Ed Lynch once paid her the compliment of saying he liked how she did the job because "she wasn't afraid to say 'no.'" Her story isn't one of the ones documents in Jean Hastings Ardell's Breaking Into Baseball, but hearing about this served as another reminder that I've got to sit down and read it after getting a copy from a friend in SABR for my birthday last month.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
I saw Pannozzo's name come up and wondered if bringing up the Boers' points is fair game or not. I'm an avid TB listener, but sometimes I question the motives, ya know?
This issue will become more of a problem if HITf/x and FIELDf/x remain unavailable to the public.

It's ok if teams are using their own algorithms to evaluate the public data, I think most people would expect that.

But when teams have the only access to data sets that are orders of magnitude larger than the public ones, the smartest blogger in the world can't compete with that in terms of creating some with similar evaluation value to propietary stuff.
Yes, it's time for a bit of sabermetric style Wikileaks. That would make for a month of scramble.
I'm not sure "troubling" is the correct word.. the industry's been going that way since front offices started hiring sabremetricians. Even those organizations not affiliated with a specific team, say BP, get accused of keeping their sabremetric tools, for example PECOTA, in a black box.
Actually, no. Within industry, the process and the products are both fully behind the curtain, leaving us to guess at their efficacy, or their impact on a team's decision tree.
True on that aspect. The production of team sabremetricians and how much it plays in the decision process is also behind the curtain except for a juicy GM quote here or there. I stand corrected.