Saturday, March 08, 2014

Pitch Framing Measurement Evolves and Avila Still Looks Good

 Alex Avila used his pitch framing skill to get more strikes in 2013.
(Photo credit: HardballTalk)

One of the most difficult things to measure in baseball is catcher defense.  In earlier posts, I have attempted to quantify this facet of the game based on how well catchers stop the running game, block pitches and avoid errors.  Those items only account for a portion of a catcher's responsibility though and say nothing about game calling or understanding of pitcher abilities and tendencies.

Another part of catcher defense that has not been measured until recent years is pitch framing or receiving.  You have probably heard broadcast analysts such as Tim McCarver talk about how catchers can get umpires to call more strikes by the way they receive the ball.  You may have scoffed at the notion that a catcher could consistently influence major league umpires to call more strikes.  Sure, umpires make mistakes but not because of the way a catcher receives the ball, Right? Actually, research suggests that catcher receiving may indeed be an important skill with more value that one might think.

Former Baseball Prospectus researcher and current Astro statistician Mike Fast tried to quantify the value of pitch framing by using Pitch f/x to determine where strikes were typically called by umpires.  He figured out which pitchers were getting more or less strikes than they should given the location of pitches.  This allowed him to isolate the effect of individual catchers on balls and strikes.  He found big differences between the best and worst catchers in the league.  Moreover, he determined that is seemed to be a skill that was consistent from year to year.

Fast determined that, from 2007-2011, part-time catcher Jose Molina got 551 more called strikes than you would expect for an average catcher.  According to former Baseball Prospectus writer and and current Rays statistician Dan Turkenkopf, turning a ball into a strike is worth about .133 runs on average.  So, Molina saved his teams an estimated 73 runs or 35 runs per 120 games with his pitch receiving skill.

How substantial is 35 runs per 120 games? That's about how many runs Blue Jays slugger Edwin Encarnacion produced with his bat last year which was a lot.  On the flip side, Fast estimated that Ryan Doumit cost the Pirates 26 runs per 120 games on balls/strikes during the same period.

One-time Baseball Prospectus analyst and current Cleveland Indians statistician Max Marchi did further work on pitch receiving making adjustments for umpire, ballpark, batter, ball-strike count, pitch location and type.  Marchi estimated that Molina saved his teams 111 runs from 2008 through early 2013, a number that caught the attention of Rays manager Joe Maddon.  Marchi also confirmed that Doumit was awful on balls/strikes.

A visual difference between Molina and Doumit can be seen in a Grantland article by Ben Lindbergh.  If you watch carefully, you'll see that there is very little movement when Molina catches a ball.  His body is still and the ball falls quietly into his mitt.  On the other hand, Doumit shows a lot more motion jerking his head and lunging at the ball as if he is reaching to pull it into the strike zone.  Lindbergh notes that this happens consistently on their pitches which could be what is influencing umpires.

The most recent work comes from Harry Pavladis and Dan Brooks at Baseball Prospectus.  They have developed new method called the Regressed Probabilistic Model (RPM) where they calculate combined probability that each pitch will be called a strike, determine the associated run value of getting a strike called on that pitch and sum over opportunities to estimate runs saved by pitch framing.

They make adjustments for ball and strike counts, pitch type, batter/pitcher handedness and pitcher differences among other things.  The best part may be that Baseball Prospectus has published the results for all catchers, something I've been hoping hoping they would do for a long time.  Their results were consistent with past studies giving further support to the notion that pitch framing is a real skill with a large impact.  According to this method, Brian McCann (127) and Jose Molina (116) have saved their teams more runs than any other catcher since 2008.  Conversely, Ryan Doumit (-124) and Gerald Laird(-83) cost their teams the most runs.

Last year, Tigers backstop Alex Avila had 5,460 opportunities to frame pitches and was able to get 72 extra strikes compared to an average catcher.  This was worth about 9 runs or one win over the course of the season which put him in the top third of major league starting catchers.  Avila has been consistently above average since 2011 saving the Tigers an estimated 17 runs during that time. 

Avila is not in the same class as catchers like Molina (24 runs saved in 2013) and Jonathan Lucroy (22), but he does seem to have a knack framing pitches and it helps the Tigers staff. It's something to watch as the 2014 season unfolds.


  1. The really neat thing is that it appears to be a repeatable skill. I didn't read the article that you cited but your summary makes sense - no body movement, ball lands softly in the glove, no lunging. It would create the impression that the ball arrived at the exact location that was intended.

    I wonder if there could be an umpire backlash against this - "Hey, Molina is making a monkey out of me. He's making me call strikes that are actually balls. Well, not any more." I doubt it. Umpires have massive egos, and you could show them Pitch f/x charts and they still wouldn't change their minds - "that was a strike", especially in their own unique proprietary strike zone.

    This is the sort of analytical breakthrough that could really change instruction for young players. I think that traditional position coaching focusses on balance, footwork and throwing release and somewhat neglects the actual art of receiving.

    It's a nice surprise to find out again that Avila rates above average here.

  2. Charles, I actually think this research could have an effect on umpires. There was an article that appeared in the HardballTimes Annual showing that umpires have been calling larger (more correct) strike zones since cameras started tracking pitches. That would explain the increase in strikeouts the past couple of years. I could see something similar coming out of the framing research.



Blog Archive


My Sabermetrics Book

My Sabermetrics Book
One of Baseball America's top ten books of 2010

Other Sabermetrics Books

Stat Counter