Discussing MLBAM's new tracking system

Bryan Cole: Let's start with what we know. MLB Advanced Media has developed a system to track pitches, batted balls, and player movement. It will be in three parks in 2014 -- Miller Park (Brewers), Target Field (Twins), and Citi Field (Mets) -- before being rolled out league-wide in 2015. From what I understood, the system has two cameras approximately 30 feet apart that work on the same principle as the human eye to judge distance; there is also a radar-based system (developed in conjunction with Trackman) to track pitches and batted balls.

And best of all, MLBAM says they plan to release this data to "all teams and the fans" once the system is installed in all ballparks by 2015. So let's dream a little: What's the first thing you're going to look at when you get your hands on this data?

Neil Weinberg: That will depend a little bit on the form the data takes, but I'm going to be really interested in two things. First, I'm really excited to get a baseline about defensive efficiency. I have a sense, based on UZR/DRS/observation, of which defenders run good routes, but I just don't know what the difference is between terrible, average, and great from an objective standpoint. Are we talking a couple of feet or twenty feet. What's the average first step compared to a great one? This is basically like learning how to watch the game for the first time because we'll be able to actually evaluate those instinctive, snap judgments we make all the time.

Second, I'm really curious to explore the variation in individual player performance. I think most people think that defense and baserunning don't go into slumps the same way offense does, but I've always kind of suspected that isn't true. The promo video showed Heyward with a 97% route efficiency rating, or whatever, but I'm going to be really interested to know how often he runs awesome routes and how often he totally blows it. That's going to be awesome.

Kevin Ruprecht: Initially, I'd agree with Neil on getting a sense of what is average, good, and bad. After gaining that baseline knowledge, I'd like to know who uses route efficiency the best to overcome deficiencies in speed, and who uses speed the most to overcome route deficiencies.

Cole: The efficiency metrics are going to be really interesting. They broke down a second play -- this one where Omar Infante gets cut down trying to score from second on a single -- and Infante's route got a 93% efficiency rating. But that turn around third looks really huge: is that fine? If he'd run a more efficient route would he have been safe instead of out by 3 feet? If I were in a front office, I'd want to look into installing this in my minor league affiliates' parks, so we could use it as a teaching tool.

Ruprecht: To figure out what's "fine", I might look at a probability distribution. For instance, what's the chance of making an out with a 94% efficiency? 95% efficiency? This would essentially have to be broken down by situation, though, which could require significant work.

Weinberg: Right, I saw someone mention this on Twitter, but this is going to make evaluating third base coaches and the decision to advance on the bases really interesting. Like you mentioned, the ability to use this as a teaching tool is going to be huge.

Cole: When does the article about how the Rays are using this info to revolutionize baserunning come out? I'll set the over/under at June 2016.

Now I'm thinking about retroactively evaluating Wendell "Windmill" Kim and it's making me depressed. Let's think happy thoughts. Let's think about the Reds' trip to Citi Field that first weekend in April. Somewhere on a server in New York is going to be detailed data on Billy Hamilton stealing bases. We won't be able to access it for a while, but still: oh man.

What's the one tool you're most looking forward to being able to measure?

Ruprecht: To start, I'm most interested in range. We can get a sense of a player's range through spray charts, but this technology could allow comparisons to be much easier.

Cole: Good point. Those spray charts are cool, but they don't give you a player's starting position. With this tracking system, you can find out how much of a defender's effectiveness can be attributed to their positioning, and how much is due to their raw athletic ability. We can see how much value a good shift adds.

Weinberg: I'm not sure how available the data is going to be, but I would love to have information on infielder release times. Range data is obviously going to be the big driver here, but knowing who gets the ball out and to the first baseman fastest would be cool.

Cole: Even if and when the data is made available, there's always a question of accessibility. At this stage, with MLBAM storing both the raw video data and the tracking information, there's something like 7 terabytes of data created per game. How do you work with something that size?

Weinberg: The 7TB number is based on every piece of data and the public doesn't need that. No one does. We only care about how far the runner went and how fast we don't need all of the data from each half second snapshot. The real challenge will be on MLBAM's end. How do they turn this raw data into usable information is real time? That's a question for someone a lot smarter than me. I'm excited to see the output, though.

Ruprecht: The size of the data generated is the biggest potential limiter of how we might glean useful insight from this technology. I don't see any way they could release all that data in a raw form, so we will be restricted to what they believe is the best method of displaying the data. I don't have a problem with this in principle, as I certainly don't have the time or computational ability to handle that much data, but we will need to be aware of how the data are presented to us.

Cole: The presentation of the data will be interesting because you don't want to overload more casual viewers with raw numbers. Even though they showed video overlays of this information at the conference, I can't imagine (say) Tim McCarver breaking down the jump Heyward got on a long fly ball on a Saturday afternoon game.

Weinberg: This is probably years away, but imagine being able to turn on "tracking" on MLB.TV and getting the information on your screen in real time. Just like anything, it's a tough balance between adding in some tidbits for casual fans while also satisfying those of us who want everything. We've started to see broadcasts show exit velocity on home runs, so maybe we're moving in the right direction.

Cole: All right, one final question: The big thing this system is missing is a name. The MLBAM people I talked to yesterday said they hadn't decided on one yet, which to my mind means that they're taking suggestions. My vote is for "BO", after the one and only Mr. Jackson. Who knows about fielding? About hitting? About baserunning? About throwing? BO knows.

Ruprecht: My vote is for "Defensive Truth", or "DT" for short. It will hopefully reveal some truths about the defensive prowess of players.

Weinberg: Kind of seems like it should be named after Mike Trout. Troutvision?

Cole: I guess it wouldn't be a BtBS article if we didn't mention Trout.

. . .

Bryan Cole is a featured writer for Beyond the Box Score, and wishes he had the tracking info for Willie Mays' basket catch in the '54 World Series. You can follow him on Twitter at @Doctor_Bryan.

Neil Weinberg is the Associate Managing Editor at Beyond The Box Score, contributor to Gammons Daily, and can also be found writing enthusiastically about the Detroit Tigers at New English D. You can follow and interact with him on Twitter at @NeilWeinberg44.

Kevin Ruprecht is a contributor for Beyond the Box Score. He also writes at Royal Stats for Everyone. You can follow him on Twitter at @KevinRuprecht.