MLBAM unveils new “player tracker” system

Open Thread: 3/1 Camp Notes
Spring Training Game Thread: Yankees @ Blue Jays

We have more ways to evaluate baseball than ever before, yet it still feels like we’ve only seen the tip of the analytical iceberg. There are still so many aspects of the game we’re unable to fully understand or accurately measure.

On Saturday, at the MIT Sloan Sports Analytics Conference in Boston, Major League Baseball Advanced Media introduced its new “player tracking” system that, basically, will measure everything that happens on the field. I’m not joking when I say everything. The ball, the fielders, the whole nine. Here’s more from Mark Newman (of, not the Yankees’ front office):

The goal is to revolutionize the way people evaluate baseball, by presenting for the first time the tools that connect all actions that happen on a field to determine how they work together. This new datastream will enable the industry to understand the whole play on the field — batting, pitching, fielding and baserunning — and enable new metrics for evaluation by clubs, scouts, players and fans.

For instance, on a brilliant, game-saving diving catch by an outfielder, this new system will let us understand what created that outcome. Was it the quickness of his first step, his acceleration? Was it his initial positioning? What if the pitcher had thrown a different pitch? Everything will be connected for the first time, providing a tool for answers to questions like this and more.

You can see the system at work in the video above, though there is much more to it than that. Only three ballparks (Miller Park, CitiField, Target Field) will have the system up and running for this season, but the goal is to having it operational in all 30 parks by 2015. I’m surprised all of this information will be publicly available, to be honest. I thought it would be kept proprietary.

Needless to say, this new system will change the way the game is evaluated. The stuff we have now is good, but it doesn’t compare to detailed information on defensive routes, first step quickness, batted ball quality, and a million other things. There has not been a reliable way to accurately measure that sorta stuff until now. It’ll be a while until we have this data for the entire league and learning how to properly use it will take even longer, but man, this new system is a gold mine.

Open Thread: 3/1 Camp Notes
Spring Training Game Thread: Yankees @ Blue Jays
  • Matt DiBari

    This is a gamechanger. Its really gonna be nice to finally have defensive analyses that doesn’t, you know, kinda suck.

    • Tanakapalooza Floozy

      Exactly! This is totally freaking awesome.

      I’ve always been highly suspect of WAR (in particular) as it’s assigned to defense-first players e.g. Gardner.

      Hopefully this non-subjective type of analysis and measurement will lead to more precise determination of defensive value, though it’ll still likely be tough (for me) to appreciate the comparative value of defensive measures to offensive ones.

  • gargoyle

    So now when s/one makes a diving catch we get to listen to saber nerds tell us how he should have caught in stride. Great.

    • dkidd

      i disagree

      (hands over lunch money)

    • Preston

      People already say these kinds of things, like Nick Swisher’s fake hustle BS, now we’ll know if it’s true or not. It always astounds me when people reject knowledge. I guess that makes me a nerd.

      • Tanakapalooza Floozy

        Well if people didn’t reject knowledge we wouldn’t have republicans.

        (Sorry. Mike. I always try to keep politics out of non-political forums but I simply couldn’t resist in this case. Feel free to delete this post if it’s troublesome).

  • Jorge Steinbrenner

    It’ll turn out, according to NBEDTEHSCRJX, that:

    1) Brett Gardner and Jacoby Ellsbury are completely different players after all
    2) Somehow, the Houston Astros won the Montero/Pineda trade
    3) Stephen Drew would actually bring on the complete collapse of the New York Yankees as we know them
    4) Cashman never really actually failed

    Bring on the new stats!

  • dkidd

    this is a breakthrough for evaluating defense, unclear what it will reveal about hitters and pitchers

    “what if the pitcher had thrown a different pitch”

    is newman saying the player tracker can explore alternate universes???

    • Nefarious

      They are utilizing an infinite improbability drive and bistro mathematics to get all those numbers.

      • Dale Mohorcic

        Don’t panic. Bring a towel. All will be well

        • vicki


    • Silvio

      Particularly since there are no alternative universes, despite what God-fearful physicists like to say.

      • Betty Lizard

        Bless your heart.

      • Brandon

        Is there a point of you making that kind of reference?

      • Dalek Jeter

        You do you, booboo.

      • Gonzo

        I would like to know about your position. I have an honest curiosity.

        • Betty Lizard

          I like the cut of your jib and I’d like to subscribe to your newsletter.

  • sevrox

    blahblahblah – how about if we just enjoy the %$#@! game?

  • CountryClub

    I’m sure this software will be handy for something (probably something we’re not even thinking about yet), but will it really shed light on defensive routes? That is one of the few things that is very clear with the naked eye.

    • stuckey

      I think there is a vast difference between saying ‘Brett Gardner appears to take good routes to balls’, and ‘Brett Gardner definitively takes a 90-something % optimal route to all balls he pursues’ and places that percentage in context vs. all other players.

      Understand that’s not an judgment on whether that’s a good or bad thing, but seems objectively clear there’s a difference.

      • Dalek Jeter

        I’m with Stuckey. It can give real, tangible evidence to statements “Brett Gardner takes good routes to the ball and therefore I can argue he will continue to be above average fielder when he loses a step.” Instead of just pointing to antiticotal evidence we can say “This defensive systems shows he has an X-speed first step and took the optimal route to the ball X% of the time.” I personally see at least some value in that.

        • whozat

          I agree, but I think there’s also an issue of scale here that a lot of people miss. What something like this gives us is the ability to have, on demand, data about routes taken by every player on every play every day over the entire history of the project. That is _amazing_.

          A scout can attest to what he saw on the day he was there. Fans can attest to what we’ve seen from the camera angle that happened to be shown, from the time that the broadcast cut to the fielder in question. How often, when watching on TV, do we miss the fielder’s first step? How often do we actually see, from a clear angle, where the fielders were positioned at the instant the pitch is thrown? How often do we see which direction they lean as the pitch is delivered? And how often does any one of us know these things for every fielder in both leagues?

          Now we get to know! THAT IS SO COOL!

          • Preston

            Yup, it’ll take time for us to know what all of these things mean and how to contextualize it, but someday soon defensive evaluation will be every bit as accurate as offensive evaluation.

            • Tanakapalooza Floozy


              And hopefully that more precise measure will bring with it a more precise determination of value as well.

        • RetroRob

          I agree, although the value to me is less in tangible evidence for us of Player X taking good routes to a ball, then what can teams and players do with the data to improve. Can a player be taught to take better routes, or is it skill?

  • Robert

    Why dont we just microchip all the players.

    Leave the game alone!!!

    • vicki

      night baseball ruined the game!

      • lightSABR

        Actually, it was the foul strike rule. Baseball’s been ruined since 1903. The National League’s been ruined even longer (big surprise there).

  • stuckey

    I’m sure guys like Billy Beane have already been using the available tools to try to gain an edge, but other than the infield shit defensive alignment has essentially remained unchanged for over a century, when the naked eye informed the optimal position for the 7 fielders behind the pitcher.

    I wonder if have a database tracking the landing spot, trajectory and speed of every ball hit will wind up radically changing defensive positioning.

    Will we find out you’re likely to do better with 5 infielders and 2 outfielders for a pitcher who historically throws a higher average of ground balls?

    Should outfielders play shallower than what’s traditional and give us a couple extra bases a game in order to cut off more hits?

    Where should a second baseman play exactly? In any situation other than a man of 1st and less than 2 out, should be be playing in short right field pretty much all the time?

    OR, will be learn the 1903 Red Stockings had it mostly right all along, despite the lack of hard data?

    • Dalek Jeter

      I personally think this will have less to do with defensive positioning (though it could have the effect of “CF’er take a step or 2 to the right against Hitter Y”) and more to do with judging defenders (especially outfielder’s) route efficiencies, their first step quickness, etc.

      • whozat

        It might illuminate that some teams consistently have poor defensive positioning, which could lead to coaching changes. Frankly, just the ability to accurately retrospect is valuable. It helps avoid the problem where you over-remember the times when you got burned on your positioning and under-remember the times when it turned a borderline play into a routine play. It can also help you separate bad process from bad outcome. I assume there are times when you position your fielders expecting that the pitcher is going to throw a breaking ball away, but instead he misses his spot and the ball gets yanked. You didn’t position the fielders badly, but they still wound up not able to make a play on the ball. Data like this allows you to see when you actually made bad decisions and try to improve them.

        • RetroRob

          I wonder about that, too. We may find that players who are supposedly good defenders, and other who are rated as below average defenders might benefit or be hurt by the team he plays on and the defensive positioning he directed to take.

          I’m sure lots of interesting things will be discovered.

    • Gonzo

      Gotta be honest, I doubt a manager would have the cajones to to something as radical as having 5 infielders anytime soon regardless of the available information. It’s the same reason why managers only use closers in a rigid way. It just makes for an easier press conference if they lose, and more importantly, an easier justification to keep your job.

      • Ed

        I’ve heard that the Pirates are focusing on extreme shifts from the minors up. I believe the GM ordered the focus system wide. Part of their recent success has come from a huge swing in their defensive efficiency. So I could see them trying something like that.

      • Preston

        Joe Maddon can’t wait to do something crazy based on this info.

  • jim p

    Will people still enjoy baseball after the next Carrington-event level solar flare wipes out the ability to go online and work out esoteric stats?

    • hogsmog

      Wow wow wow this is great; I’ve been looking forward to player tracking for a really long time. We’re finally going to get defensive stats that matter!

      Also, I was super excited to see ‘direct route vs actual route,’ and a ready-made conversion to rout efficiency. That’s certainly something that’s only ever been done by eye, in the roughest possible way (a scout can say somebody had ‘good routes’ or ‘bad routes’, and that’s pretty much the end of it).

      Looks like I might have to follow the Mets this year…

  • Wicomico Pinstripes

    “Fuck yeah”, just about sums it up for me.

  • Darren

    As long as this cockamamie system can prove once and for all that Derek Jeter is better defensively than “Bobbly” Brendan Ryan, I’m all for it!!

  • Dalek Jeter

    My immediate reaction last night and today is still “oooh neat!”

  • BFDeal

    This technology will prove what we already know…Scott Sizemore can simultaneously play 3B in Yankee Stadium and CitiField.

    • Gonzo

      That’s why they call him Scotty Quantum.

  • Deep Thoughts

    I like how they’re starting with a slow pilot program, collecting data only at the home fields of three minor league teams.

    • Deep Thoughts

      But seriously, why not give an extra year of advanced metrics to the Commissioner’s team as a going-away present, as well as his buddies the Wilpons who have more trouble remaining solvent than Lenny Dykstra?

  • BigLoving

    My first gut reaction is that I don’t like this. I’m not going to argue that this won’t change the way they evaluate talent on a more accurate basis…..but I feel like sports as a whole are being broken down like these athletes are actually stocks greedy guys on the Wall Street are dissecting to make a buck. When I go to a game I want to have a few beers, eat a couple hot dogs and relax….not get into a battle of statistics and physics.

    The world is a changing though so I guess if fans actually enjoy this I shouldn’t complain.

    • Jorge Steinbrenner

      This benefits the people who get paid to dissect performance. It does not have to affect your enjoyment of the game in any way.

    • whozat

      Do you expect teams to strap VR goggles to your head at the park and force you to watch a real-time overlay with this data? If not, what in your ballpark experience do you expect to change as a result of this?

      • lightSABR

        Can I pay for that, please?

  • George

    So if a team scores very highly in all the various numbers and calculated statistics, but don’t make the playoffs, do we get to ask for a replay and have them in as a “much wilder card”?

  • JGYank

    This is great. We were all waiting for Field/Fx and this seems just as good. This could help us get more accurate info on range, routes, reaction time/first steps, hang time, ball landing spot and the fielders distance away from the ball which is pretty much everything we need to know for defense. It could also measure base runners routes and their first step, speed, acceleration, etc so it could improve baserunning data or maybe be used to determine when to shift since we have more accurate data on ball landing spots. I’m sure it could also give data on throwing out runners and maybe even passed ball stuff as well. I’m excited to see what new data/stats comes from this.

  • vicki

    it looks like stuff they’ve been using in the nfl the last few years.

  • RetroRob

    So is MLB giving an advantage to the three teams with this system already installed?

    • pat

      I’m sure the data is available to every team. MLB wouldn’t do that.

      • Need Pitching & Hitting

        But the teams with 81 games worth of information on their own players would seem to have at least somewhat of an advantage.

        • RetroRob

          Yes, exactly. That was my point. They have full data on their own players, ballpark, etc., where the other 27 teams won’t.

          Not really sure what it means, but it seems on the surface to give an advantage to three teams.

  • Tanakapalooza Floozy

    1) If you don’t like this, I hate you. Full stop.

    2) This is completely amazing. We’ll finally have a non-subjective way to measure defense.

    3) I hope it also leads to a better understanding of defensive value relative to offensive value.

    4) I wonder whether/how route tracking and efficiency can be controlled and corrected for tag-up situations wherein a fielder may circle back behind the landing spot (a “less than perfect route” in a normal situation) in order to catch with their weight moving toward the upcoming throw.