MLB Releases More Tracking Data, Names Product
Back in March, Major League Baseball Advanced Media announced the formation of a new product that looked like the data-capturing system of our dreams. It was more concept than product, however, and 2014 was essentially going to be a year long beta test, with just three stadiums outfitted with the technology. While the system looks amazing, it hadn’t even yet been named.
That changed today, with this tweet from MLB.
The new Statcast player tracking clips will blow you away.
@YasielPuig: http://t.co/jQJJQ6wGwl
@TheCUTCH22: http://t.co/XJQ2qO2Aet
— MLB (@MLB) June 13, 2014
We welcome our new StatCast overlords.
Here are the two new plays with StatCast data that have been released.
Yasiel Puig’s insane catch.
Andrew McCutchen’s only-slightly-less insane catch.
StatCast has the potential to revolutionize the way we evaluate baseball. There’s still so much we don’t know about the system and how it’s eventually going to affect the public discourse on the game, but it’s impossible to watch these clips and not get excited. It is a certainty that the scaling and rollout of data at this size will be challenging, and we don’t actually know what kind of information will make its way into the public sector.
But we can dream. And clips like this, along with an actual name for the product, make those dreams seem ever closer to reality.
Dave is the Managing Editor of FanGraphs.
That “route efficiency %” makes me tumescent
Why is the units for acceleration seconds? Or is that the length of time they are accelerating for?
Speculating from previous clips they’ve shared, but I believe it means “seconds from first step to top speed”
That’s what I assumed – and holy hell was that a lot of hang time on that fly ball.
Of course, seconds from first step to top speed is a terrible way of measuring acceleration.
Jose Molina’s acceleration could very well be better than Yasiel Puig’s, if only because his top speed is so close to his his starting speed, of zero.
You can kind of imagine why MLB would go with that, though. I don’t imagine players accelerate smoothly directly to their top speed. I imagine they accelerate faster immediately when they start running, and then asymptotically approach their top speed.
since it gives you top speed also, that shouldn’t be an issue you. makes sense….an average speed would probably be helpful too though, that’s what really matters.
I expect the leaderboard for this stat will be headed by Jesus Montero.
OMG OMG OMG OMG OMG OMG
Kan I haz moar StatCast P0RN Plz
Seriously. This is incredible. Incredible.
This is superbly awesome. Typical, however, that as a Mets fan I had to watch these two catches again against my team. Mets pleasure is never far from Mets pain.
I’d say what’s typical is Met’s fans always finding a way to make everything about them 😉
MLBAM probably won’t release them, but I’m dying to see poor defensive plays get the StatCast treatment. Those should be much more enlightening.
Agreed. I’d like to see what an ineffecient routelooks like and what kind of percentage it’s assigned, if only for context to appreciate these 97+% routes.
See: Hunter Pence
I really doubt there would be much difference in a “bad” route because even a bad route is just point A to point B in as straight of a line as possible. Its pretty hard to deviate TOO much.
Unless the fielder does something ridiculous I doubt you would ever see it drop below, say, 90%.
But thats just my uneducated hunch.
I’m no physicist, but a bad route isn’t A–>B in a straight line, it’s A–>B in an arc, or an angulated run, or something like that. Only the 100% efficient routs are lines; I’m assuming the “route efficiency” is something like 100-[(distance run)-(length of ideal line)/(length of ideal line)].
That wouldn’t work in very unusual cases. I think it might be just (Length of Ideal Line)/(Distance run).
You obviously didn’t watch any Mariners games from 2010-2013.
good god that mccutchen efficiency
This is amazing. Looking at the numbers on those plays alone, it seems like it has the potential to take defensive statistics to the next level by basically making every play 100% objective.
I dont know if its possible to really emphasize what a huge advance this potentially is.
I wonder how many million dollars MLB is going to charge sites like Fangraphs to license their data.
need it expanded to more parks. Oakland, Houston ?
#38 Michael Morse, LF
First Step: 1.89 SEC
Acceleration: #DIV/0!
Woh. I imagine it’s only a matter of time until this is data is packaged into a product offering by someone. I’ll say 2 years before we start seeing it used on this site lol.
Can’t wait for new STAT ACRONYMS!
of course they’re both robbing mets batters…. *sigh* so it goes for a life of a mets fan.
Because they only have the system setup in like 4 parks so far.
MLB is probably the most confident in the data collected at Citi Field at this point.
Sure, but it’s not like the Mets’ outfielders have no great catches. Their CF, Juan Lagares, has surely made some great plays. He did this in the same game as that McCutchen catch (admittedly, not as impressive).
i think i just got a boner
Holy
This leaves every defender naked in front of the whole world.
It looks like the days of some of the advanced defensive stats are numbered. Why use UZR when you can just take an average of an outfielder’s route efficiency for every ball hit his direction? That kind of a number is likely to stabilize over a much shorter sample than UZR.
Of course this assumes public access to all of this data.
Because route efficiency doesn’t address range.
Like Bo Jackson said, route efficiency doesn’t measure range. I could take one step forward to make a catch, with a route efficiency of near 100% and a range of virtually 0.
People realize the numbers may not be comparable between parks? It’s easier to run in some parks than others due to the turf/grass/etc. I suspect we’ll see systemic variances between parks with the same players. They can call that system StatCast+