Q&A: Ian Levin, Mets manager of baseball analytics
by David Laurila - August 9, 2013
Like most teams, the New York Mets are into the numbers. They may not be at the forefront of analytics, but they are by no means stuck in the stone age when it comes to using data. Ian Levin is part of the organization’s saber-savvy brain trust.
Formerly the coordinator of amateur scouting, Levin currently serves as the team’s manager of baseball analytics. Last week he was part of the scouting panel at SABR 43. Afterwards, he discussed the Mets data-evaluation process.
Ian Levin on the Mets and analytics:
“I’m not sure everyone recognizes the kind of analysis we do on players. We were always an information-driven organization, and we continue to be an information-driven organization. We value scouting reports — and our scouts — and we value statistical information. We take in all information and use it the best we can to make decisions. We’re very much involved with statistical analysis and information processes.
“There is probably more of a public perception now about what we do because of Sandy Alderson and his history — his penchant for using statistics. But we’ve always done the analysis; it was just a little more under the radar before Sandy became our general manager.
“We have four people, in-house, working directly with analytics. Along with myself, there are Adam Fisher, TJ Barra and Joe Lefkowitz. Above us, Paul DePodesta is obviously involved. Everyone in the organization is understanding of metrics and uses them to different degrees. We all have different roles within the structure and work together to make decisions that help inform Sandy. As manager of baseball analytics, I oversee some of our decision-making processes and player analysis.”
On the decision-making process:
“We’re very much invested in gathering information during the decision-making process, because your decisions are dictated by the information you have. We always look back at our previous decisions and determine if they worked in our favor or not — but not to question if it was a good decision based on the result. Rather, was it a good decision based on the information we had at the time? We’re always looking to learn and get better. It’s hard to make a decision with incomplete information, so we’re all about process and information-gathering.
“There is always going to be missing information, especially when you’re dealing with a player not in your own organization. Even players in our own organization have missing pieces of information. You just do the best job you can, using your resources appropriately. Scouts get you certain information and your analysis gets you other information. You’re going to have history on players who came through the draft, and you’ll have it on international players as well. We’re always looking to find different resources to gather information we find valuable in projecting a player’s future.”
On proprietary information:
“It’s important to know your own organization as well as you possibly can; you need to know your guys better than anyone else knows them. We put a lot of time and effort into scouting and evaluating our own players. It helps us know how we stack up with everybody else, as well as project our players.
“We run all of our processes on our own players. We go out to see them play. We try to know their makeup, and we have performance metrics. Like I said, we know them better than players in other organizations. In-house, we have proprietary information on them.
“To a large degree, proprietary information is knowing subjective things. A lot of it is makeup-related, or knowing how players react in certain situations. It is information not generally captured through traditional data-capture, things like basic stats. There are also advance metrics we track on our own players. Other organizations do the same thing. Things we know about our players — and don’t share — other organizations know about their players, and don’t share.”
On video and defensive metrics:
“We use video a lot. We have video on all of our players, including our minor leaguers, as well as players throughout baseball. Video goes into every decision. If we’re looking to make a trade, we’re going to watch to see why someone is doing the things he’s doing. We combine that with our scouting reports to get a better picture of the player.
“Trying to better quantify defense is something teams are working on, and some of that information isn’t necessarily in the public domain. To find a way to value defense appropriately — within the environment of how we value it with offense and pitching — would be extremely valuable. That’s where some players get a lot of their value, and ideally you can quantify just what that value is. That’s easier said than done.
“While we’re very good at quantifying offensive performance, I think there are variables that can’t be completely measured. There are still things that aren’t quantifiable in terms of projecting how players are going to develop and perform in future situations. Fielding is the same way, but with fewer clearly measurable variables. First, we’ll need to be confident in our explaining past defensive performance properly through objective metrics. If we can do that, we might be able to get to where we want to be. Some of the information that can help is available at the major-league level but isn’t available at the minor-league level and probably isn’t going to be anytime soon. I think there is always going to be a strong need for the scouting perspective. I think scouting is the most important component of defensive metrics, and it may continue to be.”