What Do We Know About Catcher Defense? by Neil Weinberg November 14, 2014 We’ve seen some pretty revolutionary baseball research over the two decades, but until about three years ago our public estimations of catcher defense were pretty limited. We had some idea about which catchers were the best at catching base stealers, but blocking, framing, game calling, and the other nuances of the job were relative unknowns. We knew they were there, we could see them at work in individual situations, but we just didn’t have quality, public data to give us a clear pitcher of catcher defense. That’s starting to change, although we’re still a long way from home. Over the last couple of seasons, pitch framing has become a popular topic of conversation in the game with teams like the Rays, Pirates, and others seemingly targeting quality framers. We have had new metrics and seen lots of articles considering the merits of those catchers who can steal extra strikes. It’s hard to say if it’s permeated the baseball world, or just the advanced metrics/blogger world, but framing is the new “it” asset. We even saw our own Dave Cameron place a high value on catcher defense on his 2014 NL MVP ballot. Catcher defense can essentially be divided into five categories: normal fielding, pitch framing, blocking, game calling, and controlling the running game. In no area are we perfect, but there are some areas that we can evaluate better than others. Catcher defense is an evolving area of study and hot topic of conversation. Let’s briefly consider what we do and don’t know about the most indispensable position.* *-This is simply the opinion of a former catcher and shouldn’t be construed as a “fact.” Normal Fielding This is pretty straightforward as far as catcher defense is concerned. Catchers field bunts and chase pop flies and need to put tags on runners trying to score just like normal fielders. Baseball Info Solutions soaks that up in Defensive Runs Saved (DRS), and seems to do a solid job determining which catchers are capable of handling typical fielding tasks. The major problem here is that there just aren’t enough difficult “traditional” defense plays for catchers to make to really get a great handle on the distribution of performance. Ultimate Zone Rating (UZR) doesn’t even consider catchers for this reason, although there’s a chance that StatCast data might open some doors in the near future. This isn’t a buzzing area of research, but catchers are involved in standard plays in addition to all their other duties. Pitch Framing Pitch framing is ground zero for baseball analysts right now. There are dozens upon dozens of studies of pitch framing that were made possible by public PITCHf/x data, but this one from Mike Fast is one of the most prominent. He wasn’t the first or last to investigate, but that article sort of blew the doors off and gave us the idea that a catcher might be worth 3-4 wins a year based on their ability to get extra strikes. StatCorner implemented a model for capturing framing and Baseball Prospectus followed last March with an effort of their own. Efforts vary and you can easily construct a measure of extra strikes using data on FanGraphs or Baseball Savant, but the BP model, for example, takes into account the probability of the pitch being called a strike, the run value of the pitch, and attempts to control for umpire and pitcher. The research and fascination has exploded over the last couple of years. Until recently, we all knew that framing was a part of the game, but had no idea what the numbers looked like. Now, we’re in a position to say that excellent pitcher framers might be able to add a couple of wins to their team’s total by the end of a season. There’s more work to be done in perfecting some of the controls and understanding the role pitcher, but we seem to be at the point where we’re only arguing about the magnitude of the effect rather than who is very good and who is very bad. Blocking Blocking is another one of those skills that we’ve known about forever but had no real idea about how to measure until a few years ago. This groundbreaking work (pun intended) by Bojan Koprivica makes use of PITCHf/x data to estimate the probability that a pitch will get away from the catcher and then credits those catchers who keep that from happening. It’s impossible to do it justice in a summary article, but you should absolutely read it if you’re interested at all in the subject and read the Baseball Prospectus article on framing, which considers blocking as well. The model is quite impressive, but given that the talent spread is pretty limited, we’re probably not in a position to take the numbers as gospel, even if they do allow us to say that a quality catcher saves about 5 runs with his ability to block pitches. My biggest personal reservation is actually related to the next section, which is the biggest mystery: game calling. The relationship between the two is understandably foggy and impossible to strip out at the moment. Game Calling Basically, we don’t know anything. That’s hyperbole, but it’s representative of game calling’s place in the overall framework of baseball analysis. Ben Lindbergh penned a fantastic piece on the subject over the summer, but despite a very deep dive into one particular catcher’s game calling, it doesn’t leave us with a framework to truly evaluate the skill. In the same piece, Lindbergh cites data from Mitchell Litchman that indicates the spread of game calling performance over the course of a season is probably -10 to +10 runs, but the data driving that estimate isn’t publicly available. What we learn from reading that piece is that Yadier Molina seems to have a powerful effect on his pitchers compared to other Cardinals backstops, but the actual art of calling and managing a game is still quite mysterious. We don’t know anything about who really deserves credit for calling a particular pitch: the catcher, the pitcher, or the coaching staff. We don’t know anything about how often pitchers overrule catchers and choose another pitch. We don’t know about sign stealing, advanced scouting reports, or what the right pitch distribution should be in any one situation. We don’t know how often a catcher calls the right pitch and the pitcher simply executes poorly. Right now, we don’t have a game calling metric to point to because it’s really hard to develop a valid framework. Individual teams have some idea, but they have access to private information that we don’t. One of the catcher’s most important jobs is deciding which pitch the pitcher should throw and in what location, but at this moment we do not have a publicly available metric that tells us anything about that skill. Controlling the Running Game Finally, catchers prevent stolen bases and keep runners close enough to their base to limit how far they can advance on batted balls. Baseball Info Solutions has a rSB metric that captures the catcher’s ability to hold and eliminate base runners, with the goal of stripping out the role of the pitcher. Over the last year, Max Weinstein has authored two pieces (here and here) that shine light on the role of the pitcher in a big way. Certainly a great arm and quick release scare off runners and allow catchers to rack up caught stealings, but recent evidence shows that a lot of what happens on the base paths is conditional on the pitcher’s time to the plate and ability to hold runners. Ironically, the part of the game we most associate with catchers might be the part of the game in which they play the smallest role. Overview This piece isn’t meant to serve as a comprehensive reference list for every study of catcher defense conducting during the last five years. Instead, it’s meant to highlight some leading work and provide a sense of where we currently stand. We have some decent measures of all aspects of catcher defense except for game calling/leadership/managing the pitcher staff, which is probably the most interesting skill of all. Framing and blocking numbers are new, and there’s recent research which indicates our conception of stolen base prevention might be slightly off target. In general, it’s all a great start but there’s room to grow. Unlike stats like wOBA and FIP, we only have a few years of data for most of these metrics and haven’t put them through the ringer in the same way as popular hitting and pitching stats. It’s important to replicate and build on this work, but it’s also important to take stock of where we are. We knowing something about four of the five aspects of catcher defense, but we still know less about catcher defense than we do about most aspects of the game. So much of the evaluation we want to do is conditional on knowing information we simply aren’t privy to. That’s exciting if you’re curious about the game, but frustrating if you’re someone who craves precision and immediate answers.