Until about five years ago, quantifying the benefit of good defense at the catcher position was something of a chimera. Traditionalists used "Caught Stealing" as the one-stop-shop way to evaluate the defense of catchers. Spreadsheet junkies tried different variations of turning traditional catching statistics (including caught stealing) into some value of runs gained by the club. It wasn't until the mid-late 'aughts that some bright folks started using pitch f/x and retrosheet data to truly understand the defensive contribution of the catcher.
What came out of that was the growing discussion of "framing" pitches (which I am going to call "presenting" from now on, because I believe this is the term that MLB clubs have settled on) - using body language, positioning, and technique to con the umpire into calling borderline strikes instead of borderline balls. This revolutionized understanding of catcher defense, and it turns out this is the most important factor for a catcher's defense, and it's not close. The runs-saved contribution of a catcher skilled at presentation dwarfs the other aspects of defense (arm, blocked pitches, etc.) by a factor of as much as twenty times in some cases.
Today, Baseball Prospectus unveiled an update (not behind the paywall!) to their "Fielding Runs Above Average" statistic, a metric that grades player defense in terms of runs saved or lost compared to an average player with a score of zero. This update is geared particularly towards catchers and combines the runs-saved values for presentation, throwing, and blocking into one statistic.
They then added the statistic onto their Catcher Stats - full season sort-able table. This is great in that it represents perhaps the most comprehensive statistic for catcher defense publicly available. Even better, their model uses Retrosheet data going way back, and also can be applied to certain levels of the minor leagues (see the article for details), which is something of a revelation.
If one sorts the table by FRAA_ADJ (Fielding Runs Above Average adjusted for Catchers), several Astros-centric tidbits jump out immediately.
Just as we noted last year when looking at his pitch presentation statistics recorded by Stat Corner, Castro has experienced a dramatic improvement in presenting ability over the past several seasons. From Baseball Prospectus' perspective (say that a few times fast), Castro ranked as the 15th-most valuable defensive catcher in 2015, out of a sample that includes not only major leaguers, but also all AAA leagues, the Texas League, and even a smattering of Low-A. From a sample of 342 catchers who managed to catch at least one pitch, his #15 put him in the top 95th percentile in defense. He managed 11.8 FRAA_ADJ, roughly worth about two wins to the team.
If accurate (and no reason to think it's not at least in the right ballpark, as years of research have gone into this evaluative process), it easily explains the Astros' commitment to Castro in the face of continually declining offensive contribution. Castro is in the last year of his contract this year, and there is a strong case that the Astros should be considering an extension. However, there's also a strong case to not extend him, because...
It would be unjust to mention Stassi without at least noting how hopeless he has been with a bat in his hands during the past two seasons. But that might not matter much, because out of the same 342 catchers, Stassi ranked 12th, with 12.9 FRAA_ADJ, more than two standard deviations better than the average catcher.
We have been hearing good noises about Stassi's defense for a couple years, but before now had no publicly-available statistic willing to go out on a limb and quantify that for us. If Baseball Prospectus is to be believed (again, why not?) Stassi could be one of the best defensive catchers in baseball. His sample in 2014 of only 214 recorded pitches is too small to determine how much he has improved, or if he has always been this good. But his 2015 numbers are reasonably un-contestable, because presentation is a stat that stabilizes quickly due to the large number of pitches seen during a season (5,764 in 2015, in his case).
If Stassi can even manage "Brad Ausmus" production with the bat, he is a strong backup for Castro in 2016, without the Astros spending another dime on a free agent catcher, or seeking one in trade.
Wait, where did he come from? Heineman checks in at #17 on this list with 10.7 FRAA_ADJ in 3,479 pitches caught at AAA. This is exciting because it gives the Astros a trio of ML-ready (including Castro) or nearly ML-ready catchers who already have an argument as Top-20 defensive catchers who can all supply approximately two wins over an average replacement-level catcher. That's a really big deal. Heineman has not been meaningfully more successful at the plate than Stassi has, however. But in Stassi and Heineman, the Astros have upper-level depth that could step in right away, a little bit of offensive upside, and ready-made replacements if they decide to let Castro test the Free Agency waters.
Then again, Heineman's AA stats ranked him at #218 on this same list. Somebody, please explain this to me.
Once dubbed "Framin' Hank" by this website, Conger fell off a cliff in 2015. No explanation is being offered here, but after several seasons of being among the Top 5 pitch-presenting catchers in the major leagues, Conger clocked in at #71 on the 2015 list. He just ... wasn't very good. Any number of things may have caused this - injury, inattentiveness, or something else entirely, but it seems surprising and unusual (based on past years' data) to see a catcher go from one of the very best to a league-average defender over the span of a year. That Conger in 2015 did not live up to his defensive reputation and then was non-tendered by the Astros suggest that the Astros believe that his decline in skills is "for real", because prior to 2015, he was typically worth two to three wins just with his glove. Or maybe the Astros have a different way to evaluate catchers that quantifies something we aren't aware of. Or maybe they just didn't want to spend the money, since Castro is still excellent and Stassi and Heineman offer similar defensive value for a fraction of the cost. We'll probably never know, but there are several clues on this table explaining why the Astros moved on from the Conger Bot.
Other Interesting Folks
The table also offers insight into why they moved on from other catchers:
#66 - Carlos Perez (2.2 FRAA_ADJ)
#147 - Carlos Corporan (0.0 FRAA_ADJ, only 70 pitches in 2015; ranked #46 in 2014)
#150 - Roberto Pena (0.0 FRAA_ADJ, 4,751 pitches)
#155 - Trend Woodward (0.0 FRAA_ADJ, only 149 pitches)
#163 - Luis Flores (0.0 FRAA_ADJ, 548 pitches only)
#189 - Alfredo Gonzalez (-0.2 FRAA_ADJ, 2,095 pitches, AA only...hmm...)
#191 - Brett Booth (-0.2 FRAA_ADJ, 1,074 pitches)
Now, it's not surprising that younger prospects aren't ranked as well -- these players are still learning their craft. Also, one wonders from the rankings if these stats become more favorable or are biased somehow towards the Major Leagues, and then AAA, and then the lower levels, due to data availability. Another explanation is that at lower levels, pitchers are wilder, and so it limits cachers' opportunities to record strikes that would quantify their presentation skills in a way that would translate to major league projections. Actually, that makes sense, let's go with that explanation. Thank goodness for scouting!
Even with that backdrop, It was surprising to see Gonzalez rank so low, considering the Astros protected him from the Rule 5 draft on their 40-man roster this off-season. A reasonable assumption is that they see potential in his defense and work ethic, and that they like his offense (ranging between 110 to 125 wRC+ for the past few seasons) enough that they think he can round into a good defender.
Baseball Prospectus' catcher defense is the most comprehensive public quantification of this aspect of the game to date, and is a handy and trustworthy reference when evaluating catchers. The information already available goes a long way to explaining the Astros' off-season strategy in (so far) not pursuing a backup to Castro.