Keith Woolner, the Indians' Manager of Baseball Analysis and Research, was kind enough to spend an hour talking with me about any number of things recently — from his role in the organization, to thoughts on advanced metrics and the media, to a ton of other stuff. Here's the first 30 minutes or so (coming in at a incredibly reasonable 3220 words) of our talk. Part II coming tomorrow, or whenever I finish transcribing it.
VG: First, where did you grow up?
KW: I was born in New Hampshire. And I lived there until I was about ten years old, then we moved to Orlando, Florida, and I graduated high school there, so I lived the rest of my youth in the Orlando area.
VG: Who was your team?
KW: The Boston Red Sox. Until I got the job here I was a lifelong Red Sox fan. But, your loyalties shift remarkably quick.
VG: Yeah, I read in the BP chat that it came really soon.
KW: I started with the Indians in 2007 and a few months later we're in the playoffs against the Red Sox. There was no sort of divided loyalties or anything. It was clear which side I wanted.
VG: Ok, so when did you first start getting into the advanced statistics?
KW: I was a fan from a pretty young age, but when we moved to Florida I got away from it for awhile. I started getting back into it in college when I went to MIT. I could see the lights from Fenway shining in my windows at night, and just having that sort of access rekindled a passion for baseball. And right about that time there was an emerging online community that was talking about baseball and some weird guy named Bill James who I never heard of. So I started following those discussions as a reader and got interested and said , Hey, I wonder if I could come up with some of my own stuff, and I published that stuff on my own web page.
Eventually it got the attention of the guys who had recently started Prospectus so they invited me to come on as a writer. So I ended up working with them as a hobby or second job for about ten years. Ended up co-authoring a bunch of books, built up a lot of the statistics portion of the web site, that led to when I had a chance to look around for my next career, I thought well, let me talk to some teams and see if there's any interest in a guy like me, and the Indians responded and here I am.
VG: Let's go back a second. What was your main job.
KW: For most of my life I was paying the bills in the software industry in a technical support role and after five or six years of that, transitioned into product marketing/product management sort of role which was less figuring out what code to write but what to tell the coders to write. Figure out what products to build. So it was still a lot of interaction with engineering and technical folks, but not doing the work so much as handling it.
VG: What was your initial interaction with the Indians?
KW: There had been a round of layoffs at my previous pre-baseball employer, I said I've been doing this baseball thing for awhile, there've been a couple of people who managed to get in the front office.
VG: How many teams had a position like yours as you were exploring this?
KW: A handful. It's a little hard to say because teams aren't always forthcoming and not all of them have come out of the online baseball community like Prospectus. Probably no more than a half dozen. Certainly fewer than that that had come from my background.
But I thought, a couple of people have had some success, including some people from Prospectus, so I used it as a way to say, Hey, do you know any teams that might be looking. They put out some feelers. And Mike Chernoff here responded and said, yeah, we'd like to talk to him.
VG: So when you showed up who did you sit down with and what was the interview process like?
KW: It was Mike Chernoff and Chris Antonetti. I met a few other people throughout the day, but primarily it was the two of them. And it was a pretty thorough interview, probably as thorough as any that I had in the software world. And it covered a lot of areas, both what you might expect sort of quantitative or statistical capabilities and what kind of approaches did I think I could bring to certain problems they were facing. But also how they were kind of assessing my personality and what would the fit with the team be, and could I interact with people who weren't from a software or technical sort of background. So, there were trying to evaluate the fit all the way around. It was a big commitment for them and us.
VG: What was their department or analytic capabilities were before you showed up?
KW: They did not have a department. They had certainly been aware of a lot of the research that was going on online, they were familiar with some of my work. There had been some thought of maybe bringing in an intern or grad student to help with some of the problems they either didn't have time for or didn't quite know how to solve. So, it was kind of fortuitous — they were thinking this is the time we need to bring someone in, and I was looking wondering if anyone needed help.
VG: Reading interviews with Shapiro and Antonetti, they seem pretty in-tune with this stuff, and I guess I was wondering if that had been going on for awhile or if they were pushed even more in that direction after you arrived.
KW: I think that for both of them, they both recognize there was an opportunity in exploiting this sort of analytical approaches. Alongside, of course, scouting and everything that's traditionally been done. There was this other avenue of information that could help make better decisions. So, they didn't need much convincing that there was some value in doing this. It was more the question of whether I was the right guy to do this or what structure and did it warrant bringing someone in, hiring them and relocating their family.
VG: So, you were the guy when you started. What kind of staff do you have now?
KW: I have an intern who's been with us for several months, but at present, I'm the only full time analytical baseball research person.
VG: What about the sorts of data they were collecting. What level of collection were they at? Did you come in and recommend tracking a whole bunch of new stuff?
KW: For the most part, they had recognized that, to some degree, you try to collect as much information as you can, because even if you don't know how to use it today, at some point in the future it might be useful. They had developed, prior to my arrival, an internal system called Diamond View that's pretty well known in the industry as being a good way to consolidate and view all the information they've collected. It gives them easy access to scouting reports, to stats, to whatever they need to look at.
So, there wasn't so much when I came in and said we're missing a lot data. It was more, we have this data, we're not applying it the best we could.
VG: Was there anything in the last year or so that you suggested they track that they weren't two years ago?
KW: I can't think of anything specific that I came in and said we need to do this and we're not doing it right now. Generally, when they realize there's more data out there, it's been long before I got here, they recognized they needed to acquire it and it had value.
VG: IN the BP Chat, you mention the difference doing Prospects and what you do now is the access to the sheer amount of data.
KW: You can use data sort of on a simple reporting basis. If I want to know what a given player, I can look at the scouting reports I can look at the medical information and the accumulated stats, but there are other applications you can use it for. Not so much for one player, but what are the trends we can tease out by looking at a lot of players in a lot of ways. What are the methods and applications we can bring to bear on that that maybe we haven't thought of before, that we can borrow from another field.
VG: While I look for my next question, are you and your family enjoying Cleveland?
KW: I came from North Carolina and lived in California prior to that. When you tell people you're moving to Cleveland they get this funny look on their face. But we have enjoyed it much more than we expected to. The winters are aren't as bad as we feared, maybe it's because we moved to the west side instead of the east side. It's been a very pleasant surprise. We really enjoy the city.
VG: The data that's collected on the major league level, is that same down the line through the minor league system?
KW: To a large degree we have the same kind of data available. In some cases there's some differences in the granularity in the data. One example is the Pitch FX data. Major league parks have the cameras in place to track the ball in flight, there's a lot of analytical studies that are being done with that. That's not available in all the minor league parks. We try to have it as complete as we can all the way down, but there are some differences.
VG: What is your relationship in how decisions are made in the front office?
KW: They have been very open to the input that I would have to give. They ask for my opinion on a number of moves, potential trades, offers, we're thinking about promoting a guy, contracts. I don't think there's any case I can point to and say that the feedback that I gave them was the sole piece of evidence that convinced them to do something. In a lot of cases, I don't think in any of their decisions that you can point to one thing, to the exclusion of everything else, they're going to rely on. They look at everything. My role is to be a source of information and to be as expert in the area that I'm responsible for as I can to make sure they have the best information and they know both the strengths and the limitations of that data.
VG: Has that changed at all since you started? Are they more receptive than when you started?
KW: I think there are two things at work. One is, as you work with someone over a longer period of time, you get more comfortable, you know what each person's biases are or how to best communicate your point. And the second thing is the longer I've worked here the more tools and research and familiarity with our systems I have, so I can do more applicable more in depth types of work, which in turn, is more useful to them.
VG: You said in your chat with BP, you're mainly involved in player acquisition and your input in possible moves.
KW: What I think I said, when you look at decisions and in-game moves, technical moves, how often to bunt, bring in a closer, the things you can in the game to improve your managing vs. the player acquisition and evaluation of talent, I've done work in both areas, but the majority of it has been on the player evaluation/projection side.
VG: In terms of lineups, I know it's a very small difference in the matter of a season, is that sort of information available to the people who make those decisions — say, hypothetically, whether to move someone from shortstop to third base, or to move someone from the leadoff spot to batting third in the lineup?
KW: That information is available. I only have experience with one front office, but from what I'm told and believe to be true, there is more receptivity here among the coaching staff, Eric and so on, to whatever is going to help them get the right results and make the best decisions. If that comes from a statistical approach, that's OK. If it comes from a psychological evaluation, that's fine too. But, so often think about or hear the old baseball guys who would dismiss anything a guy like me would have to say out of hand, that has not happened. Not to my face.
I've had conversation with Eric and other people in the organization,and they've been at least willing to listen, and that's a big part of the battle — just having the conversation.
VG: Have they asked specifically for information regarding those hypothetical moves I just mentioned?
KW: Umm. They have asked for input in difference situations. I'll just leave it at that.
VG: I figured I was going to get a bunch of those murky answers.
VG: How did the work you were doing for Prospectus differ from what you do now?
KW: There is, first of all, the work that I do now matters in a way it didn't for Prospectus. For Prospectus, if people were reading what I had to say, right or wrong, it was a win. Wins are very different when you're actually playing games.
Another way it differs, is there's less interest or emphasis on evaluating the past. When you're writing for a baseball site, you can write about the MVP race in a certain year, and there's people interested in that, and it's interesting research. But in running a team, the past, in some sense is only useful in that it informs the decisions we're making in the present about the future.
VG: So what are more of the predictive stuff that you look at, beside strikeout rate?
KW: (laughs) Um. I think that the kinds of things we look at are not going to be surprising, at least to your other population of readers. Age, strikeout rate, contact rate —the kind of things you would expect to go into that model do go into it.
VG: Part of the attraction of the online community is sharing ideas, dissecting ideas, improve ideas. Is that something you miss.
KW: Yeah, absolutely. A lot of the work I have done that I think is really cool, I want to say, "Look what I found." This is a different way of doing something than what is being done in the community. I think it would be, at a personal level, sort of gratifying to be out there, because that's what I did for ten plus years. But, having it be a little bit of a secret that hopefully will be a competitive advantage to maybe help win the World Series is pretty cool too.
That was a tradeoff I went into with my eyes open. I don't regret it, but there's times when it would be nice to be able to share.
VG: Let's talk about the media. I was watching Baseball Tonight last night, and Nate Silver's there, and Nate Silver's in Esquire, and Prospectus is on ESPN, and I think they had WHIP as a stat the other night, and OPS shows up everywhere. Are you happy with how things are slowly progressing in the media towards using this stuff?
KW: Yea. If you look at the past ten years, the fact that you can have a guy like Nate Silver on a major sports network, and he can talk credibly about the game, when you can get OPS, or even VORP has shown up once ore twice on those programs, that's a big win. That's progress. That's getting these ideas out there and making them more mainstream. That's, from a community perspective, that's one of the nice things about it. One of the benefits you get from sharing your work is that you see it accepted, you see it reach a broader audience than you ever thought possible.
VG: OPS seems pretty understandable to the general public. What do you think might be next, that the public could understand easily and something that's incredibly useful?
KW: OPS obviously is one. That's a good question.
I think that stats that are represented on a familiar scale or easily understand scale are at an advantage at reaching the mainstream fan. That's one of the reasons something like EqA has ben successful, not because in and of itself it measures something, or that an EqA of .270 is .27 of something, it's on a scale that's familiar and natural. So, if you look at something like the DIPS ERA or FIP, the fact that it's expressed on a scale that looks like the ERA that people are familiar with will ease it's acceptance.
VG: Instead of like ERA+, like what the hell does 117 mean?
KW: Right, I still have trouble with that (laughing).
And I think that one of the most powerful things that could be out there that could be done, is the idea as runs being the sort of currency by which you purchase wins.And the rough rule of thumb that ten runs, ten marginal runs, equals one win. That's a surprisingly simple and powerful notion, because you can take any aspect of the game and if you can translate it into what it's effect on the final score, you can then project that over the course of a season and say this creates this many runs which creates this many wins, as an expectation. I don't think we're there yet, but I think that's an idea which is, maybe it's a little wish casting, but that's the idea that I wish would become mainstream next.
VG: Where do you think that pressure is going to come from? Newspapers are shutting down and laying off writers and all the old writers who don't like the advanced stats will eventually retire, and there's more blogs about this stuff. Is it going to be newspapers trying to attract that audience, or is the audience saying, we're just not going to read you if you don't include this?
KW: I think those are two aspects of the same phenomenon. I don't think those are two distinct causes. One's the supply and one's the demand. But I think that it is going to be, as you get generation fans who have grown up with a different set of experiences in terms of what they've read and what they know, just the fact they've been exposed to computers their whole life.
VG: Growing up with FireJoeMorgan instead of Joe Morgan.
KW: Right, at least didn't watch Joe Morgan play. They're going to have different expectations about the news that they consume and I think that whatever media outlets end up surviving the next 25 years are going to respond to that. They're going to make that available. Now, whether it's going to be broad based enough to where you can expect it to show up across all relevant baseball media.
VG: Like a box score?
KW: Right, I don't know what the standard box score of the next 25 years is going to look like yet.