New defensive metric??

by Cardsray on Aug 11, 2011 1:16 PM EDT

(disclaimer: OBP, OPS/OPS+, WHIP, and FIP are about the peak of my sabermetric understanding, so if what I'm about to discuss is an actual metric out there somwhere, then sorry, I'll delete the fanpost)

I was watching the game last night against the Brewers and saw a pitch from Westbrook go WAY outside. Yadi lunged about 4 feet to his right and made an extremely good catch with a runner on third. That got me thinking about a possible "defensive stat" for catchers. I'd appreciate any discussion from the saber-smart folks around VEB. The idea I have is sort of an UZR/fielding bible +/- for catchers.

Scenario: With a runner on base (any base), if there is a pitch that is anything but right around the plate, there is a chance of the runner advancing (i.e. past ball, wild pitch).

Let's say that over the course of a full season a catcher has 100 of these scenarios (It would obviously be a lot more, but let's make the math easy here). 100 times during the season there was a runner on base and a pitch that was either in the dirt, way outside/inside, or high to where the catcher would be unable to stop it from his normal rested crouch position.

My question to the people of VEB is: is there a way (with Pitch FX, or whatever tool) to easily find out how many times this happened and what the results were??

Using the scenario above, let's hypothesize that Yadier Molina had 100 of these instances last year. 92 times he stopped/smothered the ball and the runner did not advance (good). 8 times the pitch got away and the runner advanced (bad). By that logic, he would have had a 92% efficiency, or +84 rating (+92 times he stopped it/-8 times he didn't), or however the stat could be termed.

Anyway, anyone smarter than I care to help me out in understanding this? It would basically be a way to compare catchers around the league purely by their ability to stop/smother an out of control pitch when it actually matters (with a runner on).

Thanks in advance.

Tweet Comment 11 comments | 0 recs |

Do you like this story?

Comments

Not sure but I believe I read something about a stat to that affect earlier on

Though I believe it also rated Ryan Hanigan above Yadi, so I’m not sure it was the most accurate

Someday when I'm old and have grandchildren, I'll sit them down and say, "I got to see Ryan Theriot play SS." And they'll reply, "Oh, I'm so sorry you had to experience that."

Currently residing on the Free Tyler Greene bandwagon

by cardinalswsbound on Aug 11, 2011 2:35 PM EDT reply actions

Im sure

There’s some stat out there that shows some sort of catcher defensive metric. My idea would take into account the play I mentioned with molina. It was a terrible pitch, way outside but nothing happened (no runner advance), so it will just go in the record book as a “ball”, instead of “catcher made a great play that would have definitely resulted in a passed ball”. That’s kind of what I’m getting at.

by Cardsray on Aug 11, 2011 3:30 PM EDT up reply actions

I don't see why this wouldn't work

The only problem is that Pitch f/x would be pretty inadequate for judging how hard the ball was to smother. There might be some correlation for the odds of a runner advancing based on the distance of the ball from home plate, but I would bet it would be very small.

Secretary of WAR and defense (Tyler Greene Fanclub). PUT TYLER ON THE GREENE.

by vivaelpujols on Aug 11, 2011 4:44 PM EDT reply actions

Yeah

I think you’d have to have a model based around at least five things:

1) distance from plate
2) speed of pitch
3) ball in dirt?
4) ball thrown to the side with catcher’s dominant hand?
5) LH/RH batter?

They say that it's never too late, but you don't get any younger...

by Valatan on Aug 12, 2011 12:42 PM EDT up reply actions

And then, considering the nonlinearity of the variables

That model would be pretty weak.

Secretary of WAR and defense (Tyler Greene Fanclub). PUT TYLER ON THE GREENE.

by vivaelpujols on Aug 12, 2011 4:39 PM EDT up reply actions

Yep.

They say that it's never too late, but you don't get any younger...

by Valatan on Aug 14, 2011 5:36 PM EDT up reply actions

Of course, you could probably collapse some of these into single variables

By, say, using the LH/RH batter, and side of pitch to create an ‘effective distance from plate’, but that has it’s own problems (model dependence).

They say that it's never too late, but you don't get any younger...

by Valatan on Aug 14, 2011 5:42 PM EDT up reply actions

Pretty much any regression based model you do with Pitch f/x is going to be pretty shitty

I’ve found that similarity scores are the best way to use the data, but those probably wouldn’t help with this type of catcher metric.

Secretary of WAR and defense (Tyler Greene Fanclub). PUT TYLER ON THE GREENE.

by vivaelpujols on Aug 14, 2011 11:07 PM EDT up reply actions

This isn't exactly what you describe, but it may give similar information.

Over at The Hardball Times, there was a piece by JT Jordan called Catcher Defense 2010.

"I'm gonna throw the nastiest curveball I have ever thrown...if he hits it, I'll tip my cap, but if not we're going to the Series."

--Adam Wainwright on the final pitch of the 2006 NLCS

by bgh on Aug 12, 2011 12:00 PM EDT reply actions

I think it's an interesting concept

but there are too many variables. I think pitchers have a lot more to do with passed balls and wild pitches than catchers do. This idea kind of reminds me of the error, in a sense that just because a catcher “misses a catch” doesn’t mean that he didn’t play it well. Maybe most catchers would not even reach the ball. Also, if a catcher catches for a pitching staff that is more “wild” than another, one would think he would be graded poorly, since he would allow more passed balls/wild pitches.

From what I’ve seen I think this is probably the worst aspect of yadi’s defensive game. The guy has a great arm but I don’t think he blocks pitches very well. I remember Matheney being far more talented when it came to blocking pitches, but he didn’t have the arm yadi does.

check out VEB on facebook...just search groups for Viva El Birdos

by Dttl89 on Aug 12, 2011 4:30 PM EDT reply actions

When it comes to the catcher position, there are always going to be too many variables to quantify

I love sabermetrics as much as the next guy, but I’m going to choose to ignore them in regards to the catcher. Sometimes it’s best to keep some things mysterious.

by peppermartin on Aug 13, 2011 7:54 AM EDT reply actions