Unveiled at the 2014 Sloan Analytics Conference, MLBAM’s player tracking data has been in full swing (no pun intended) this season, at 3 stadiums (Citi Field, Miller Park, and Target Field), and is planned to be utilized at all major league stadiums by Opening Day 2015.

In order to collect the data, special cameras are located around the ballpark and measure various metrics involving the positioning of a player and of the ball. For base runners, this includes first step reaction timing when stealing; max speed/acceleration; lead off the bag; etc. For fielders, similar metrics are used but perhaps thought of differently: first step reaction timing to get to the ball, route efficiency, distance traveled, max speed/acceleration, etc. Then, when a catcher attempts to throw out a runner or an outfielder makes a throw home, arm strength/speed, quickness of release time, and accuracy are measured. Finally, we can see more advanced speed and angle metrics for batters and pitchers than were currently already available, such as pitch exit speed off the bat, pitcher extension, spin rate, etc. Each of these data points are definitely useful in evaluating such talents as speed, defensive prowess, and strength in ways more direct than stats like stolen bases, defensive runs saved, or slugging percentage. While it may be visually apparent which players are better at these skills, there is now a measurable way to evaluate and compare players – useful when many players appear to be visually similar in this regard.

Currently, MLBAM makes some of the data publicly available through their video player. A quick search will lead you to many of these plays from this season. Although we are not fully aware of how teams are using this data to evaluate or coach players, it’s not hard to envision how this data can be used. What interests me the most is the idea that a player might have a ton of plays you might see on a highlight reel, but it does not necessarily mean he is a better defensive player than someone who does not. If an outfielder is able to get to a distant spot where a fly ball is hit much more quickly and efficiently than another player, he might position himself easily under it and make the catch look easy – where a slower/more inefficient fielder might make a spectacular diving play. The latter play looks flashier and more impressive; but I would rather have a guy on my team who I know is more likely to get to where the ball is quickly. These are the kinds of skills that should become easier to evaluate. A player that can make the most difficult plays look routine is probably more likely the better defensive player.

Because you are able to tell how hard and where a ball is hit with more precision than in the past, this can be used to gauge that a hitter might be hitting well, but getting some bad luck. Perhaps, for example, a player is hitting the ball very hard lately, but still getting out. In general, the harder you hit a ball, the more likely you are to get on base. So, a stat like how often you hit the ball hard is very useful for gauging power, while how often you get a hit when hitting the ball hard is useful to see if the player is getting unlucky (such as hitting it hard right to a defensive player). On the flip side, knowing how hard or where a batter hits a ball can tremendously help with defensive placement of fielders. This is already apparent in “playing the shift” against certain batters who don’t pull the ball often, but there will be more statistical ground to stand on, to have a more concrete shifting strategy, when this data is more available. More examples of how this type of data is being used in sports can be found on Mark Simon’s blog on ESPN, as he often publishes player/team stats on these metrics.