Story URL: http://news.medill.northwestern.edu/chicago/news.aspx?id=184885
Story Retrieval Date: 9/30/2014 11:01:04 PM CST
Technology has enabled statheads and sabermatricians to usurp scouts as the all-powerful evaluators of hitters with their new statistics and rigorous analysis, but number crunchers have never been able to accurately track fielders.
That is, until now.
“I feel pretty confident making a comparison between two hitters,” said Sean Forman, creator and operator of the statistical database Baseball-Reference.com. “But when it comes to fielding there is more subjectivity [and] the data is not as clean so it becomes more difficult to analyze the players.”
Clean, quantifiable data is on the way as Sportvision— the Chicago-based data content provider that developed the virtual yellow 1st and Ten Line in football and the K-Zone in baseball— is implementing a system called FIELDf/x in major league ballparks that digitally records the position of all players and hit balls in real time.
“The problem is that a lot of advanced defensive statistics have been based upon observation,” said Mike Jakob, president of Sportvision. “With the FIELDf/x system, we are going to be gathering over a million data points on each game. What that will do is give us the ability to truly start to mine the data and figure out what are the relevant statistics.”
Sabermetricians – baseball analysts who rely on objective evidence – are getting their algorithms ready because FIELDf/x could be the key to the first objective metric for measuring defense.
“It will end a lot of arguments that we’re having right now because of the fielding data,” said Forman, who has a doctorate in applied mathematics from the University of Iowa. “You’ll be able to actually show the distance traveled to field every ground ball. Having data for that would just be tremendous.”
Sportvision has already given fans and analysts advanced technologies for tracking pitchers and hitters. Its PITCHf/x system, now a staple on major league broadcasts, uses three cameras and a central tracking system to digitally record the full trajectory – with accuracy within an inch – of live baseballs as they are thrown to the plate. Its HITf/x system tracks and records data from the trajectory of hit baseballs.
The FIELDf/x technology, first implemented at AT&T Park in San Francisco last season, uses three additional cameras to track each fielder’s every move.
It can show that the reason a fly ball dropped five feet in
front of the left fielder is because the fielder did not move until 1.5 seconds
after the ball was hit. Unless you are at the ballpark and are carefully
watching that player’s every move, you would not be able to notice this subtle
but crucial fielding deficiency.
“And it’s not just range questions,” Forman said. “It’ll be stuff like, ‘What’s the pop time for every catcher?’ It will be rigorous and in theory it will be consistent from player to player, game to game, so it will be much more accurate.”
The rise of sabermetrics has led to a more complex analysis of baseball through the creation of new statistics. Until now, most of those measurements are a product of number crunching as opposed to the mapping of physical data.
The Sportvision technologies, above and beyond allowing for the teasing of useful statistical data, provide an accurate representation of the physics of baseball as it happens on the field, Jakob said. Consequently, the FIELDf/x system will challenge the analyst community as it presents them with more objective data than ever.
“I don’t think the sabermetricians realize yet just how different it’s going to be analyzing this data,” Forman said. “You’re talking about like 20 billion cells of data just for a single season. Analyzing that is going to take some new tools and new analysis that we don’t have at the moment.”
Sportvision has offered limited data sets to teams and some baseball analysts. Jakob said the company is taking its time before deciding how and to what extent it will release the data to teams, sabermetricians and fans. Above all, Sportvision wants casual fans to have these technologies at their disposal without inundating them with data.
“Over time you want to put consumers more in control of how much they see and when they see it,” Jakob said. “Certainly the broadcasters will selectively use it, but we have certainly seen a pretty strong appetite from the fans to control what they see.”
For sabermetricians such as Forman, more data is always better. So although they do not know how they will handle the ocean of FIELDf/x data, they are excited to start wading through it.
“I don’t know yet what it’s going to be like to deal with a table that has a billion rows,” Forman said. “But people will start with the simplest question and then work from there and start publishing what they have done. And people will build on it like any other scientific process."