Future of Defensive Statistics and Scouting

[crewcrazy]

Brian and Bill mentioned this briefly during tonight's game, and I thought it was an interesting topic. Apparently they've installed overhead cameras at AT&T Park in San Francisco that, when used in conjunction with computer software, can help better evaluate defensive talent. The New York Times recently did an article on the new system and includes screenshots and videos of what the software looks like in action.

 

Since most fielding metrics out there can still be controversial, it should be interesting to see what comes of this new system. With a computer program being able to peg out exact coordinates on the field, there should be less guesswork involved. I'm personally looking forward to showing data like this to Yankees fans I know to prove that Derek Jeter isn't nearly as good as they think he is. http://forum.brewerfan.net/images/smilies/wink.gif

[brewtown82]

This is interesting, but without more teams being able to use it, I don't think we'll get its true impact any time soon.

[pebadger]

It's supposed to be in every park next season.

[logan82]

It would help eliminate or reduce the main problem with most defensive metrics which is their failure to account for defensive positioning before the pitch.

[kramnoj]

Thanks for the link, that was a great read. I was surprised to learn that the company doing all the work to install the equipment doesn't really know how to use the data it will be collecting, but I guess that's not their thing.

 

It will be interesting to see how expensive they will make the data for the public, and who will be able to afford it. And of course, assuming that some will be able to afford it, how will they decide to use the data, and how intensive will that work be? How does this affect Baseball Info Solutions? This is just a guess on my part, but they seem like a company big enough to take on the task of using this data. It seems like it would be too much for people to do anything with this on a part time basis. Of course, the biggest part is figuring out what to do with the data. Once that has been done, I guess the rest is just maintenance.

[topper09er]

Maybe in the future the gold glove will actually be a meaninful award that is not based on how well you hit.

[crewcrazy]

Maybe in the future the gold glove will actually be a meaninful award that is not based on how well you hit.

It'd be nice, but unfortunately I think it's more likely that we get a flurry of sportswriters with votes writing columns that say things like "I don't need some kind of computer to tell me who's a good defender" if they're proven wrong, or "It's a completely useless system because it tells me what I already know" if the data ends up supporting their opinions.

 

While the Gold Glove voting will probably continue to stink, at least if the public can get their hands on this data we'll have a better idea of who's a good defender and who's living off their reputation.

[topper09er]

The potential for this is pretty cool. It would nice to see some stats which could break defense down a little bit into its parts; range, fielding, throwing and awareness. Range would account for how many balls you can get to, like lateral range for a SS (from up the middle to in the hole) or cutting of balls in the OF (like if recognizing that if you are playing deep it is not as impressive to cut off the same ball as if you were playing more in) or getting good jumps in the OF. Fielding would account for how well you cleanly handle the balls you get to. Throwing could even account for things like how often a SS hits the 1B in the chest compared to high throws or throws to the side or in the dirt. Awareness could measure how often you throw to the right bases or times when the cut off man should have cut the ball. Im not sure how you would add them together for an overall defensive rating it would be cool to rank all the SS based on how far to either side they can go to get a ball that took 1 second to get to them.

[logan82]

They already do a fairly good job of defining how well a player fields balls they get to. It is easy to tell if a guy gets an out when they get to a ball. The big problem has always been range. The argument almost always comes down to who is better the guy not making mistakes and a guy with far superior range making more fielding mistakes(not errors those are a horrid way to look at defense). The problem is trying to sort out who would convert more balls in play into outs.

[topper09er]

True, but fielding percentage counts errors as fielding and throwing errors combined into one stat. Also, for example, when Furcal threw that low throw in the 1st inning on Saturday, you could rate that as a bad throw by Furcal but also just as bad fielding by Russel Martin. Its easy to say that Prince costs an out by not being able to scoop a throw but you could also account for times when the SS cant scoop a relay throw from the OF.

[brewers19]

Screw the statistical potential, that's just plain fun to watch. It's like RBI Baseball, but without the hive-minded outfielders.

[RightFieldCoder]

This might also be used to create baserunning stats. We can watchthe 2nd runner scoring. I'm sure smarter people would be able to come up with a way to measure their running statistically.

[NievesNoNO]

Good read and interesting stuff. I actually think we're kind of on the cusp of a transitional period in the philsophy of the game with regards to defense. There are stats that have been in place for a few years now such as zone rating that are widely available and I'm sure have been tracked longer than fans have had access to them. I don't know the others or really even the specifics on that particular one but they're there. I also recall reading a SI article earlier this season on how teams were starting to focus on defense more. They highlighted the Phillies and Rays as well as the Mariners. Anyone else recall that article? Anyway, teams and GMs seem to be more aware of the value of good defense and I think those elite defenders will start to be paid more in line with their ball mashing no defense playing counter parts. The technology is certainly there to do it.

[TooLiveBrew]

Good read and interesting stuff. I actually think we're kind of on the cusp of a transitional period in the philsophy of the game with regards to defense. There are stats that have been in place for a few years now such as zone rating that are widely available and I'm sure have been tracked longer than fans have had access to them.

 

Great point, NievesNoNO. I've been looking forward to the ways in which defensive statistics & data collection will evolve over the course of the next 10-20 years, and I think these cameras will be a huge first step.

[pebadger]

I think they have been using a similar system on a limited basis to track soccer players for a while over in European leagues and in international matches. They can determine exactly how far guys have run in a game, and can get the times and distances of the more explosive runs that guys will make down the field in an attacking situation. I think they also track the ball itself, with respect to time in the air versus on the ground, distances of crossses into the area and other such stuff.

 

With so much data available through video analysis sytems like these, I can't help but think that all sports will undergo a fair degree of scrutiny and change in terms of the intellectual approach to the game.