THE BOOK--Playing The Percentages In Baseball

Here’s the link:
http://seattletimes.nwsource.com/html/marinersblog/2011361700_geoff_baker_live_with_bill_jam.html

I thought this exchange about catcher defense at the 7:57 mark was interesting.

GB: How precise is it though?  That’s a good question that somebody brings up.  As far as things like catcher defense, I mean, it’s not a very precise field, let’s say.

BJ: Catcher defense is really hard.  We cannot and we’re not on the cusp of being able to measure catcher defense very accurately.  There’s so much we don’t know, and maybe...I would bet 40 years from now we won’t be too far ahead of where we are now on catcher defense, although we may know things we don’t know now.

GB: So what do teams do?  Do they just throw up their hands and say, well, I mean yeah, we can’t measure this?  How do they go about picking somebody then? Based on...do you go by gut instinct? Do you have to think you know what direction you’re heading in, or what do you do?

BJ: You can have a lot of knowledge that’s not measured. I mean, with the Red Sox we know that Jason Varitek spends a great deal of time
studying videotape, and we know that he knows exactly what every hitter’s weaknesses are, how likely any hitter is to get himself out on any pitch he can throw. We know that. What we don’t know is exactly what difference it makes.  So
you have to rely on scouts; you have to rely on your coaches; you have to rely on your instincts about those kind of issues.  But then 35 years ago we had to rely on those people about everything, so we’re making progress.

I feel like we are on the cusp of being able to measure catcher defense accurately, probably within the next five years.

"I feel like we are on the cusp of being able to measure catcher defense accurately, probably within the next five years.”

I agree and I thought the same thing when I heard it.

So why weren’t the Mariners a true 75 win team last season?  This is an honest question.

Jimbo:

http://www.insidethebook.com/ee/index.php/site/comments/pythag_records/

Mike & MGL:

Measure it how?  Are you talking about teasing out the proper coefficients for a formula that uses box score values?  Something that measures location and time parameters such as pitch tracking, catcher motion tracking, tracking the catcher’s glove, etc.?  Some combination of all that, or something that doesn’t currently exist even in concept?  What?

I’m asking because I’m interested to hear why you think it’s within 5 years.  Do you think we’re 20% (or whatever %) of the way there now on a path that you think leads to the answer?  Or are you just confident that something will turn up with enough smart people digging, and with technology providing ever more information?

I’m not in the contrary position to yours, I don’t really have an “outlook” on catcher defense right now.  I’m interested, though, and wondering if you could elaborate on what you’re both thinking on this…

Greg, Pitch f/x mainly for blocking pitches and framing.  Better methods, like WOWY for separating catcher throwing skill from pitchers having slow or fact delivery.  Eventually, base stealers and catcher throws will be part of the “f/x” right? 

And then pitch f/x for seeing how different catchers call games differently with the same pitcher on the mound.

That is about it for catchers right, other than his ability to keep the pitcher focused and calm during adversity.

As far as the M’s not being a true 75-win team, we can never know what true talent a team is, but while pythag record likely reflects a team’s true talent better than their actual record, the teams underlying numbers (offensive wOBA, UZR, team FIP, and baserunning) are a MUCH better reflection of team’s true talent than pythag record.  That is BP’s third order wins, essentially.  For that, I think the M’s were an 83 win team, or something like that…

Greg/6, yes to what MGL said.  I’d also say that I think we’re already quite far along on catcher defense if you take the leading edge of the work on each facet.  That stuff hasn’t been developed and refined to the point that it’s Fangraphs-ready yet, but it will get there soon. 

I hope in five years we’ll have a much better handle on game calling with a wealth of PITCHf/x data and a few more years to study it.