Mike Fast and his amazing insights about catchers and the way they glom pitches come to you in an afternoon chat.
Mike Fast: Hello, fellow peeples of the Internets! Thanks for all the support and feedback on my recent article on catchers. Feel free to ask questions about that or about other topics.
Nick (Jakarta): Is there a chance that all the new and insightful research you and others are doing on catcher fielding is going to dramatically change the way we view certain players? For instance, will we start to view someone who we thought as an average catcher and player as an elite one after all the research is done?
Mike Fast: I don't know that it will be that dramatic, but an extra win per season is pretty significant. It's an important part of the total catcher package, but not necessarily more important than hitting.
Dave (Chicago): When the late A's owner Walter Haas gave the Giants the territorial rights for the San Jose-area, did he get anything in return?
Mike Fast: As far as I know, he did not.
R.A. Wagman (Section 203): Mike - do you know of any research about the effect of the pickoff throw (beyond actually catching the runner) on an offense? It seems to me to be on of the few things that occurs on the field of play that is not yet tracked. Thanks.
Mike Fast: Which Section 203 is that? I wanna be at the ballpark, too!
I am not aware of any research on catchers preventing baserunner advances beyond the direct effects on SB/CS/PO. Just because I'm not aware of it doesn't mean it hasn't been done.
On a related note, it's definitely an interest of mine to figure out how to separate hit-and-run plays from CS plays. I'm not convinced we do a very good job of measuring the impact of catchers on the running game, yet. For one thing, pitchers make a big difference, but in addition to that, there are potential effects on non-SB/CS plays.
Lucas Apostoleris (Amherst, MA): Mike, would you be able to talk a bit about what you think the next steps in evaluating catcher defense are?
Mike Fast: Well, catcher framing could certainly use some firming up in terms of getting more accuracy on the numbers. I don't think we'll see a big departure from where Max and I have found them, but we're not at the last word on that topic yet by any means.
I want to extend the work that Dan Turkenkopf, Dave Allen, and Max have done on measuring pitch blocking using PITCHf/x. I'd like to see how much accuracy it buys us compared to the Retrosheet-data work done by Justin Merry, Matt Klaassen, Lee Panas, and others.
I mentioned the running game above.
Pitch calling is one arena where I don't have a good handle yet on what to do next. I actually had some good conversations with my brother this weekend about that, but I still don't know exactly where to go. Sean Smith's work on catcher ERA definitely points us in the direction of believing there is a significant pitch calling effect, especially when combined with our work on catcher framing, but I don't know that we're anywhere near quantifying with precision yet.
JD Sussman (Long Island): Mike, have you had a chance to analyze Wilson Ramos behind the dish? Narratives aside, if he is as good as he has looked, he should be a serious NL ROY candidate as a lg average offensive player with plus defense.
Mike Fast: I have not done any video analysis of Ramos. The numbers in my study (which you may have seen if you looked at the Google spreadsheet that I linked) show Ramos to be among the best catchers at getting extra strikes, but in less than a full season of duty, fwiw.
ttt (Manhattan (yay work!)): I just wanted to say I really enjoyed the article on framing pitches. I always wondered what the impact was of catchers framing (or just attempting to frame) pitches.
Mike Fast: Thanks! I really enjoyed researching, writing, and sharing it.
Tom Hanrahan (Lexington Park MD): Mike, has anyone to your knowledge done a PBP study of results (ball, strike, batted ball) from pitches thrown after a pitcher has shaken off >=1 catcher signals? Might this be a valuable piece of info? Pure "% of time shaken off" figures for catchers, possibly in comparison to teammate catchers and/or in tandem with pitchers, could also be helpful?
Mike Fast: I'm not aware of anyone doing that kind of study. I will say that any study that requires viewing video always makes it tough to accumulate any reasonable sample size.
That's one reason it was very helpful for this catcher study to trim down the sample size of pitches I needed to view by using PITCHf/x plate location data. It's still time intensive to view video even then, but it's a lot easier to get somewhere than it is when you're essentially viewing pitches at random.
Brandon (Annapolis): Have you been contacted by any MLB front office people Re: the catcher piece with beefs about how the data may indict their catcher?
Mike Fast: Not exactly, no. I've had some very helpful comments and criticisms from baseball insiders, but nobody has disagreed with the study on the sort of basis that you're asking about. The feedback from all around has been amazingly positive, even among people who have pointed out things that I missed or that they think I missed.
Lucas Apostoleris (Amherst, MA): Based on pure stuff/dominance/command (whatever strikes you), which pitchers are you most looking forward to watching this postseason?
Mike Fast: Kenley Jansen is the guy that has piqued my interest most recently, but I'll have all offseason to look at him, I suppose.
Of the pitchers who look playoff-bound, Craig Kimbrel is just amazing. I'm curious about Verlander's BABIP this year. Doug Fister is an interesting story. I've wanted to look at Alexi Ogando's switch to the rotation. Koji Uehara fascinates me, with his splitter and high flyball/popup rate. Josh Collmenter and Ian Kennedy are interesting.
Of course there's no way I'll end up analyzing all or even most of those guys, but those are a few names who grab my attention.
Dan Turkenkopf (Saratoga Springs, NY): For something totally not catcher related, do you have any thoughts on Matt Lentzner's look at plate discipline from the Sportsvision summit this year?
Mike Fast: It's definitely an approach that I want to evaluate. For those of you who didn't follow the summit, Matt looked at plate discipline by dividing the strike zone into three areas: always-strike, borderline, and always-ball. It seems like a superior approach over a binary ball/strike division.
I wish I had something more insightful to add, but at this point I haven't had a chance to play with the data that way, so I don't really know much more.
Jquinton82 (NY): Do you see Jesus Montero catching enough games next year to keep his catcher eligibility in most leagues?
Mike Fast: All indications are that Montero's defense at catcher is not very good. Still, the bar for keeping catcher eligibility in most leagues is what, five games in season, right? So I would think he would cross that bar with ease.
Of course, Joe Girardi is really the one you want answering that question, not me.
The only helpful piece of data I have to offer is that Montero graded out very poorly in a VERY small sample of catcher framing data from September. Well, maybe that's not so helpful after all. It's such a small sample as to be practically meaningless.
hotstatrat (Toronto): PITCH/fx judges strikes and balls as it crosses the front of the plate. Because of this, umpires probably disagree with PITCH/fx more often with curve ball pitchers than with fastball pitchers. You normalized the pitchers each catcher faced in your pitch framing study, which would effectively normalize pitches that might miss the front of the plate but catch the back of the plate (and vice versa). True?
Mike Fast: The PITCHf/x data does report the location of the pitch as it crosses the front of home plate. It also includes enough information that you can reconstruct the location of the pitch at any other point on the trajectory, though. So the analyst is not restricted to judging balls and strikes at the front of the plate. In practice, the pitches don't move a whole lot between the front and back of the plate--a half an inch or an inch--but it's worth considering when you're trying to fine tune something like this.
I've tried in the past to figure out how umpires call the strike zone for curveballs. It's not easy to separate the effect of the pitch type from the effect of the ball-strike count (mainly) and other factors. Curveballs often get thrown with two strikes when umpires seem to shrink the zone a bit. Pitch types, particularly sliders and changeups, are thrown with differing frequencies to different edges of the plates depending on batter and pitcher handedness. So it's not exactly easy to disentangle the effect of pitch type on the strike zone.
However, my bigger concern with measuring how many extra strikes catchers get on curveballs is that receiving mechanics for curveballs are very different than receiving mechanics for other pitch types. Catchers tend to set their feet wider apart and position themselves to be ready to block a ball in the dirt. That's not necessarily conducive to receiving a pitch with quiet and stable mechanics, but I haven't looked into the statistics for framing on curveballs specifically to see what effect it has.
Scoresheetwiz (www.scoresheetwiz.com): If a sim game such as Scoresheet wanted to use catcher framing as part of their game - do you or BP have any plans to make your findings available to such a group in the near future? I'm not sure what kind of format they would need it in - or what sort of proprietary control anyone has on it.
Mike Fast: We're working on integrating catcher framing measures into our statistics here at BP. No target date for release at this point; it's still very much in the development stage as far as automating things goes.
As to licensing those stats, I don't know how that's done. Probably a good conversation to take to email if you would like.
Geopipp (Victoria ): Mike,
I believe the Jays have an interesting problem in the near future. Travis D'anard and J.P. Arencibia will bring challenges to the franchise as I think they will have to choose between them. I don't think that J.P. is willing to play second fiddle. Travis just continues to impress as he moves through the baseball ranks. Thoughts?
Mike Fast: J.P. Arencibia is one of those guys whose name my brain/fingers always want to misspell. I start typing Arecnibia...
I don't know enough about D'arnaud to give an intelligent answer, but Arencibia's bat is pretty good.
dianagramr (NYC): Hi Mike ... great work on that research. One question though. Did you make any adjustments for catchers who need to handle knuckleballers (Wakefield) or others with bad control (Burnett)?
Mike Fast: Thanks, Diana! To a first order, my adjustment for pitchers should adjust for catchers who had to handle knuckleballers or pitchers with bad control.
In terms of second-order refinements, the issue with knuckleballers definitely attracts my attention. From what I have observed, knuckleball catchers don't tend to set a target for the pitchers as such. They rest their glove on their knee and stab at the pitch when it comes. That wouldn't seem to be very good for getting extra strike calls, but in fact, Wakefield in particular and Dickey to some extent get bigger strike zones than average. So perhaps umpires are as confused as everyone else by the knuckleball and give the pitcher/catcher the benefit of the doubt.
If I had a catcher who primarily caught the knuckleball and no other pitchers, I'd be concerned that he would not be adequately measured by this technique simply because I'm not sure how/why umpires call the zone they do for knucklers. I'm not so concerned that this technique would be so far off on knucklers, though, that it would invalidate the results for catchers who caught the knuckleball and plenty of other pitchers. It's definitely worth additional research, though.
Dan Turkenkopf (Saratoga Springs, NY): Mike,
Any intention to write a followup to your list of "don'ts" with PITCHf/x? (Especially highlighting don't use per pitch actual run values?)
Mike Fast: Dan's referring to this piece that I wrote at the Hardball Times:
The Internet cried a little when you wrote that on it
I don't have a follow-up piece planned. It's always less rewarding, at least initially, to write about things other people are doing wrong. I do feel like the Internet Cried piece has served as a good reference, though.
I agree with your concern about how actual per-pitch run values are used. The problem with me writing anything about that is that I don't think we know enough about how game theory really operates in baseball in the pitcher-batter confrontation. Sky Andrecheck had a great piece on that at Baseball Analysts before the Indians hired him. Someone else, whose name escapes me at the moment, did some theoretical work on Brad Lidge and slider/fastball usage. But we really know so little about how game theory in fact operates that we don't know if actual pitch run values are close to being true or complete fiction or where on the continuum in between those two poles.
My biggest peeve with PITCHf/x usage continues to be people that latch onto changes of a couple inches in pitch movement as being significant for a pitcher without examining the likely confounding factors in the data.
Jquinton82 (NY): What do you see Granderson putting up next year? Obviously he's worked Long to make adjustments against lefties... is there any adjustments you see leftie pitchers making to him?
Mike Fast: He's pulling the ball a lot more this year against LHP, which would seem to indicate that his mechanical adjustments were effective.
I suppose LHP could try to pitch him more outside to keep him from pulling the ball. He hasn't seem ed vulnerable to that this year as he's laid off pitches off the plate outside with great discipline.
The whole area of batter and pitcher adjustments is one that is ripe for analysis as we gather up multiple years of PITCHf/x data.
Jquinton82 (NY): One last question and I know it may be early for this one... Any hitters or pitchers you like a breakout candidates next year?
Mike Fast: Hm, I like the look of Jerome Williams in a small sample this year. I think Fister and McCarthy may be the real deal, though they're not exactly breakout candidates any more. I like Brian Duensing coming into this season, and still do.
Dave Robertson and Kenley Jansen are two names that will surprise no one, but they could end up in bigger roles. In that vein, Greg Holland for the Royals, Vinnie Pestano for the Indians. And in the this-guys-stuff-intrigues-me-but-his-wildness-scares-me category, plus he has a cool name: Fautino de los Santos.
a-nathan (Urbana IL): Mike...followup to Jquinton82: what about Carl Crawford. Without having done any analysis beyond eyeballing what I see on TV, it appears his extreme open stance makes it difficult to handle a pitch breaking away from him on the outer half. Have you looked into this with PITCHf/x data?
Mike Fast: Hi, Alan! Interesting question. I have not looked into it, though one of the things I would really like to study is how batter stance affects results and/or how the pitchers pitch to them. That's an interesting observation, and worth some study.
Clearly, I need a wealthy benefactor so I can sit around and crunch data and watch baseball video all day!
Andrew (A meeting from hell): What do you think the Cubs do at 1b next year? As well, is rebuilding the best approach, or should they try to regroup for a run?
Mike Fast: Sorry about the meeting from hell. Meetings are one of my least favorite activities, and long, boring ones, well...
I follow the AL a lot closer than the NL, so I'll have to admit I don't know enough to give a good answer to your question. Harry Pavlidis (@harrypav on Twitter) or Colin Wyers (@cwyers) would be much better equipped.
It looks like Carlos Pena was a decent one-year signing, but if the Cubs can't compete, it wouldn't be worth trying to re-up him. I don't know if Pena has any desire to stay.
I would think that the Cubs potentially could be competitive in the next year or two, though probably not favorites, but coming from a Royals fan, take that with a grain of salt. I don't know what the Cubbies have in the minors at first base or whom they could potentially move there.
Bill Simmons (LA): Thoughts on grantland.com so far?
Mike Fast: I've read a few articles that I liked, by Rany Jazayerli and Jonah Keri. Of course, I liked their writing before they signed up with Grantland. Other than that, I haven't read much there to give me a more informed opinion.
Marissa (OC): Brett Lawrie Career All-Star appearances, 2.5, Over or Under?
Mike Fast: I'm going to say "under" simply because that's the safe bet when he currently has zero, not because I have a crystal ball. Also because 18 HR in Las Vegas doesn't make me believe his short season power in Toronto is all its cracked up to be.
Travis Snider (Seattle): Is it too early to label me a bust?
Mike Fast: You're 23 with 800-some plate appearances in the majors. Yeah, it's probably too early to label you a bust. I'm not buying your stock simply because the walks and strikeouts aren't great, and you haven't displayed enough power to really grab my interest. But you're young enough and have displayed enough potential that someone ought to give you another chance.
Mike Fast: Thanks everyone for your questions. There were a number of good things to think about, and I always enjoy getting new ideas to ponder.