Five Star Defensive Plays

[AlwaysinModeration]

As many are likely aware, MLB rolled out a new feature this year via Statcast, "Catch Probability."

http://m.mlb.com/news/article/217802340/statcast-introduces-catch-probability-for-2017/

 

In a nutshell, catches are ranked based on how likely your average position player would have made a catch on a similar ball.  A percentage is assigned; what percentage of players make that catch?  Good catches are given stars according to this scale:

 

0 to 25 percent --- 5 Star play *****
26 to 50 percent -- 4 Star play ****
51 to 75 percent -- 3 Star play ***
76 to 90 percent -- 2 Star play **
91 to 95 percent -- 1 Star play *

 

So far this season, both Buxton and Kepler have gotten five star plays.  But is MLB keeping track of how many 5s, 4s, 3s, etc each player makes?  I can't seem to find it. 

 

By the way, I got to thinking about this because of Sano's outstanding play on the Gomez bunt last night.  That had to have been a <5% play, although the Statcast stat doesn't apply to grounders, only fly-outs. (I guess it makes sense because there are two players involved, harder to parse.) 

 

That Sano play was amazing.

 

Anyway, the new feature is great, but it would be good to be able to see some more data on the starred plays.

 

 

[ashbury]

I've seen this before but never had looked at the details. Apparently the two variables on the graph, the distance needed to cover and the time to do it, are the entire essence of the metric. If it doesn't take account of the wind, and maybe glare from the sun and other sundry factors, then there will remain gray areas, which is why the stars are defined as percentages of such plays made - not just for the universe of players trying to make that play, but the same player under different conditions. So it's not an absolute metric, but nevertheless it looks very useful, and I'm glad if MLB is embracing something like it.

[Doomtints]

This is a 1 star stat designed to give dialog to the nightly highlight reels on ESPN and the MLB network.  It probably helps their editors hunt for exciting plays throughout endless hours of footage as well. Now they can just look at the stat sheet for 5 star plays and cue them up.

[bluechipper]

That's the thing I don't get with Statcast. They show all of these stats on MLB Network, but then they only have a few categories listed on the MLB website.

 

I'd like to see the fastest times from home to first or the fastest throws from the outfield or things like that. Hopefully they put more stats like those up sometime soon.

[AlwaysinModeration]

That's the thing I don't get with Statcast. They show all of these stats on MLB Network, but then they only have a few categories listed on the MLB website.

 

I'd like to see the fastest times from home to first or the fastest throws from the outfield or things like that. Hopefully they put more stats like those up sometime soon.

Agreed. They must have all of the data, just aren't posting it.

 

I have to think seeing all of the 5-stars, 4-stars, 3-stars etc would make it more possible to see which players are having the biggest impact in the outfield.

[Eris]

 

As many are likely aware, MLB rolled out a new feature this year via Statcast, "Catch Probability."

http://m.mlb.com/news/article/217802340/statcast-introduces-catch-probability-for-2017/

 

 

 

Anyway, the new feature is great, but it would be good to be able to see some more data on the starred plays.

 

https://baseballsavant.mlb.com/catch_probability_leaderboard?min=0&year=2017

 

On the home page, https://baseballsavant.mlb.com, it is found under Applications (right most menu on the top tool bar).  You can also access 2015 and 2016 data from this menu.

 

I posted in the forum about my analysis of the 2016 data about 1 month ago.  I transferred the data to a spreadsheat and also looked at % of balls caught.  It gives a better picture as elite outfielders catch about 80% of their opportunities.

http://twinsdaily.com/topic/25407-statcast-and-twins-of-defense/?do=findComment&comment=596761

 

[Respy]

They said they are working on infield data.  They said it's a lot harder, but they'll get there.  I would assume the difficulty is that the balls take different bounces and spins (as opposed to a flyball which just has a hangtime and location).

[AlwaysinModeration]

https://baseballsavant.mlb.com/catch_probability_leaderboard?min=0&year=2017

 

On the home page, https://baseballsavant.mlb.com, it is found under Applications (right most menu on the top tool bar). You can also access 2015 and 2016 data from this menu.

 

I posted in the forum about my analysis of the 2016 data about 1 month ago. I transferred the data to a spreadsheat and also looked at % of balls caught. It gives a better picture as elite outfielders catch about 80% of their opportunities.

http://twinsdaily.com/topic/25407-statcast-and-twins-of-defense/?do=findComment&comment=596761

Great analysis, thanks for posting the link and your work.

[Oldgoat_MN]

 

https://baseballsavant.mlb.com/catch_probability_leaderboard?min=0&year=2017

 

On the home page, https://baseballsavant.mlb.com, it is found under Applications (right most menu on the top tool bar).  You can also access 2015 and 2016 data from this menu.

 

I posted in the forum about my analysis of the 2016 data about 1 month ago.  I transferred the data to a spreadsheat and also looked at % of balls caught.  It gives a better picture as elite outfielders catch about 80% of their opportunities.

http://twinsdaily.com/topic/25407-statcast-and-twins-of-defense/?do=findComment&comment=596761

It won't let me LIKE this, but I do.