The next evolutionary leap in stats
A company in Australia is deploying a personal GPS unit on Aussie football players to track their movements during a game.
This device is exactly what’s needed to bring baseball defensive stats to the next level. Well, not exactly for baseball, as the strap-on contraption would be a little unwieldly for baseball players. I can’t imagine it would take much to modify the device to put it in the sole of a player’s shoe, for example.
While there are ever advancing analytical models to better quantify a player’s defensive ability (or inability), the biggest sticking point is quality data to study. Defensive statistical analysis is based largely on inferences or data that are not exactly a great metric of ability (see error, earned run, putout, assist). Some video and software companies are attempting to rectify this by throwing college students at thousands of hours of video to manually record player starting position, end position, speed, acceleration, turning, leaping, etc.
Sorry, that’s an Industrial Revolution solution. It’s a bit too Mechanical Turk for me. It’s an understandable temporary manual solution until a proper Information Age solution became available. And now there is one.
Ideally, we’d also get these GPS units inside the baseballs too, so we could finally record acceleration off the bat, deceleration in the air, speed, and flight path. Combine that with high speed cameras and optical sensors to track the spin of a pitch and it’s contact with the bat and we will open a whole new era of baseball statistical analysis. Add in real-time data those in the dugout can use to make adjustments and you have a potent leg up. Of course, that would require a lot of computing power at the park.
Don’t think for a second that someone like Billy Beane and the A’s or Mark Shapiro and the Indians haven’t already envisioned this future.
Shoot, we might finally be able to get an accurate measurement of home run balls while we’re at it too!
In a recent interview, Bill James was asked about the future of statistical analysis in sports. His point about improving leagues is valid. But there is still a lot to learn about the specific causes of baseball defensive ability, such as speed, positioning, etc. that will let us move away from inferences based on results. Once we collect these data, it will open a whole new world of analysis.
November 1st, 2007 at 00:52
We most certainly have the capability now to be doing this. Road Runners and bicyclers have devices strapped to their shoes for timing purposes. The most likely way in which it will be applied is by baseball teams tracking the effectiveness of their infield and outfield positioning. Strap it to the shoes and away you go. It wouldn’t get in the way of anything. People can be placed within inches of their most effective position to defend a hitter vs. a certain pitcher. Then there’s old curmudgeons like me who think that their is too much thinking going on sometimes on the field. Swing at strikes and pipe it if ya gotta!!!