The WAR against Molina (by Chris Tyrrell)

Credit: Cut4 of MLB.com

Credit: Cut4 of MLB.com

This is a guest post by Chris Tyrrell. You can find him at @LuckySTLFan on Twitter

If you ask the average BBWAA writer to break out the chalk and explain every aspect considered in WAR (rWAR or bWAR), not a single one of them could write a single equation and include all variables.  Not one.  For example, BsR, a stat provided on fangraphs is defined as the following:

A*B/(B + C) + D

Where A can mean H + W – HR, H + W or it can mean HBP – HR – .5*IW.  B can be (1.4*TB – .6*H – 3*HR + .1*W)*1.02 or it can be (1.4*TB – .6*H – 3*HR + .1*(W + HBP – IW) + .9*(SB – CS – GDP))*1.1

BsR further broken down here.

If you dig around the variables in fWAR, you can spend days taking notes and making spreadsheets and still find yourself lost with no greater understanding of the game when you finally give up.  This is not to say that WAR is not a great idea, because it is.  The problem with applying VALUE is that in anything in life, value is based on the eye of the beholder.  Mike Trout is the golden boy for fWAR and rWAR.  He seems to lead the world across the board each year.  Yet his team hasn’t made the playoffs with one of the league’s highest payrolls in that span.  How much VALUE can one centerfielder and one spot in the lineup truly add to a team?

So how do I intend to apply value to Yadier Molina? Simple truth is I can’t.  There is no one stat on catchers that I can justify weighing more than others, or that of other position players.  But I can point out the work of those who have tried, and then add a very blatant oversight in the game of sabermetrics just to show you how ridiculous the idea of WAR being objective when it so obviously fails to account for variables it never thought to account for.

Catcher ERA might be a function of pitching.  Ok, it IS a function of pitching.  But pitch framing changing catcher ERA is measurable and can objectively be argued.  Now this argument falls for one of the fallacies of all sabermetrics.  Instead of just being a statistic which should be compared between catchers, it tries to associate real world results as value.  Meaning RUNS SAVED, WINS vs. the average, something no stat should ever try to do.   You can’t assign runs to an individual in a game that requires other people on base during your hits or driving you in and have that equation be consistent over any period of time.  You can’t assume some coefficient that works one year to loosely associate WINS when every player faces different batters with different umpires calling different strike zones each night.  What you can do is index each stat based on its own merit without warping it into runs or wins and use that measure to compare players to one another.

That same issue occurs in the following study by Jared Cross of ESPN on pitch framing, based on the work of John Roegele’s strike zone analysis here.

It also tries to assign WINS to pitch framing.  How do you accurately translate strikes to outs, outs to runs saved, runs saved to wins?  You can assign a formula, and call it objective because it’s math.  That doesn’t make it accurate.  Same issue exists with WAR.

So based on this study, Jonathon Lucroy would have had a higher WAR than McCutchen.  And so would Molina.  They saved their teams 4-8 wins roughly just by expanding the pitching zone with their framing.  Think it doesn’t matter?  When Molina was on the DL this year with an injured knee, not a single Cardinals starter got through 7 innings not named Wainwright.  Even Wainwright never got through the 8th inning.  Every single starter was averaging significantly more pitches per inning.  There are two possible reasons.  1. Pitch framing failed to translate close pitches to strikes and more balls means longer at bats, more walks, and more pitches per inning.  2. The backup catchers Cruz and Johnson were simply not calling as aggressive of a game. “Pitch to contact” is a sure fire way to get pitch counts down.

So statistically speaking, pitch framing alone can make up the WINS according to studies derived from the same math that WAR is based on.  Molina’s framing scientifically has been proven to create 19 more strikes a game than the average catcher using pitchFX.  That’s an extra inning pitched by his starters.  That’s the difference between 180 IP and 210IP over the course of a season, and you know how much value is weighted on innings when debating similar pitchers with a 30 inning difference for a season.  That value is easily recognized for one pitcher.  Now apply that value to 5 pitchers.

Mike Current and Chad McEvoy wrote a great story on the value of base stealing, and its contribution to the statistic WAR and how it contributes to runs in both college and in the MLB.  I won’t go into UZR and DRS as these speak for themselves on Molina’s behalf.  But it should be noted that while stolen bases help the base runner’s value in sabermetrics, the catcher’s prevention of stolen bases is completely ignored.

For example, the Run Expectancy Matrix created by Baseball Prospectus reveals that a runner on first base with no one out is worth approximately 0.864 runs. A successful steal of second base would raise that figure to 1.173. However, a failed stolen base attempt drops that number to 0.270. In this example, the loss is nearly two times the gain.

So a stolen base in this case gains 0.309 runs if successful and loses 0.594 runs if unsuccessful.  Now it varies with how many runners on base, how many outs, etc.  But it is always worth more to the defensive team to catch a runner stealing than it is to the offensive team to steal the base.  Using a median value of 1.6 times caught stealing = 1 time successful steal, and 0.700 runs per base runner, Molina’s 26 stolen bases allowed are equal to 18.2 runs against.  Molina’s 20 runners caught stealing are equal to 22.4 runs for.  So his DRS should actually directly translate to 4.2 more runs saved.  And his WAR should be adjusted accordingly.  This is the hard evidence.  If we were playing poker, this would be calling a 10 dollar bet in a 20 dollar pot with a flush draw because your 10 dollars now wins 30 dollars, or 3 to 1 odds on a 38% chance of winning.  Those are known quantities that absolutely exist.  But a great poker player would tell you that the implied odds are even greater.  It is assumed even MORE money would go into the pot later in the hand and if you hit your flush draw, you stand to make that money as well.

Here are the implied odds of Molina and base running defense:  No one makes it a habit of running on Molina (outside perhaps Billy Hamilton who can run on anyone he pleases). Alden Gonzalez, of MLB.com, touches on the effect of having Molina on your team and base runners here.

He paraphrased two Cardinal starters:

Joe Kelly: Hey Adam, when’s the last time somebody tried to steal off you?

Adam Wainwright: I don’t know if somebody’s tried to steal off me the last three or four years of my career; I can’t even remember it.

Kelly: Yeah, same here.

Now my argument can be highlighted by this conjecture, but mind you this is NOT my argument.  I only want to use real world tangible numbers.  So let’s prove their pithy self-patting conversation with real world statistics.

The Cardinals allowed 39 stolen bases this year.  Again please note that 13 or 33% of these are against Cruz and Johnson, not Molina.  We will even ignore the 13 allowed by the other catchers to offset any error inherently associated with implied odds, or implied effects on run value here.  League average stolen bases allowed this year was 89.76 according to the following source:

http://www.sportingcharts.com/mlb/stats/team-stolen-bases-allowed-per-game/2013/

(…and some basic math of course)

This means the Cardinals have allowed, as a team, 50 less stolen bases than the average team.  MLB teams averaged 123 stolen base attempts during the 2013 regular season.  Yet only attempted 46 steals against Molina all year long.   Given Molina’s number of games at catcher, he should have seen 98.7 stolen base attempts based on league average.  So this ‘implies’ his presence behind the plate resulted in 52 less stolen base attempts, and 41 less stolen bases weighed by games played / 162.  Now using the same math from before that would have been 41 * 0.700 runs for steals or 28.7 runs minus 0.700 runs *11 caught stealing * 1.6 (weighted for more run value catching vs. successful steal as previously discussed) or 12.3 runs which comes to 15.5 runs saved simply because teams are scared to run on Molina.

Now we can argue runs saved in the stolen base department having a lot to do with pitcher delivery to the plate.  You hear this a lot from proponents of Buster Posey citing the Giants pitchers are very slow to the plate.   Even if half of the credit goes to the Cardinals pitchers, and half to the catcher, then you can statistically add another 7.75 runs to Molina’s DRS of 12 (fangraphs) just for the fear of other teams when he is behind the plate.

Just as a side note, the Cardinals had 1.10 higher ERA when Molina wasn’t catching.  It’s a small sample size, but a full run is extremely telling no matter how much you want to bury your head in the sand on catcher value.  I could argue semantics, or other lofty catcher value with no real statistical information like how he “handles the staff”.  When Shelby Miller recorded 27 straight outs, he said “I say it time and time again, what Yadi calls, I throw,” Miller said. “He was calling the right thing all night. He’s done a terrific job all year and he’s helping me out tremendously. I’m happy he’s my catcher, that’s for sure.”

You can google Molina and “called a good game,” and there will be thousands of stories full of such quotes.  Again, this is not quantitative and impossible to argue.  So it’s a good sentiment but doesn’t add to this argument objectively.  Unfortunately catcher’s ERA could only truly reflect this aspect of his game if you had 1500 games caught by Yadi, and 1500 games caught by his backups to definitely prove a gap in the quality of pitching based on the catcher.  One thing that can be looked at loosely is lifetime ERA of players throughout the minors and then their vast improvement when called up to the Cardinals and being caught by Yadi.  Even this is subjugated on the likes of pitching gurus Dave Duncan and Lilliquist influences at pitching coaches.  It is an impressive improvement, but again can’t be argued conclusively.

For the hopelessly romantic, Molina handled a staff with no less than 10 rookies this year. Carlos Martinez, a prospect noted for a lack of control and inconsistency, drastically improves when Yadi comes to the mound to settle him down in big spots.  Yadi can speed up a game, slow down a game, watch the batter’s eyes and posture, and move the defense around based on his homework of each batter.  The intangibles argument is lengthy and can go on seemingly forever.  Even polls of random ESPN sports fans largely agree he could be considered the best defensive catcher of all time.

http://espn.go.com/sportsnation/poll/conversation/_/id/3615869

Writers argue for and against that with Johnny Bench being his only substantial competition:

http://espn.go.com/blog/sweetspot/post/_/id/34292/molinas-defense-grows-to-legend-status

So his offense was good this year.  His defense was typical Yadi and off the charts good as seen by his 12.0 DRS.  His control of the run game conservatively saved 10 more runs (15.5 in raw data).  His pitch framing alone was worth another 4 wins above replacement based on some translations you may or may not agree with.  What is undeniable is he saved his pitchers 19 strikes a game in pitch framing.  As stated earlier, that’s about an inning a game, or 130 innings over the course of a season.  He was essentially worth the same number of innings pitched as a 5th starter.  And he’s not even a pitcher.

I don’t claim to define catcher value here.  If I spent the next 4 years working through every possible measurable quantity discussed in these last few pages, I might be able to come up with a stat about as legitimate as WAR, but I’d also probably emerge from the mountains of box scores with a PHD in statistics.

While some would argue this is insanity because it means catchers are the most valuable player on every single team if this were true, I can easily put your minds at rest.  These stats are not based on runs or wins vs. any other position player, these are just Molina vs. the average catcher.  For everything he is the best at as a catcher: Hitting, blocking balls in the dirt, throwing out runners, picking off runners, fielding his position, blocking the plate, pitch framing, calling a game; a typical catcher does not have positive value at all of these.  Most break even, with positive value at some of these, and negative value at others.

And while most members of the BBWA have no problem admitting that Yadier Molina unquestionably is the best at most of these aspects of being a catcher, it seems that 28 of 30 voters did not feel like applying value to most of them, just hitting, and to a much lesser extent fielding.

It’s not that catchers are the most valuable player on every team, it’s that Molina happens to be the most valuable catcher at almost every aspect of catching.  When combined with the bare minimum metrics people seem to care about these days (hitting, WAR and to a much lesser extent UZR and DRS), these other aspects which are much harder to define make Molina the most valuable player in all of baseball.  Even if I hit all of the derived runs and pitches and base runners wiped out by Molina with a coefficient of 0.50, claiming a 50% margin of error, he still tops the league in WAR if you add the outs added and runs saved into the equation.  Just one stat alone, pitch framing, can be shown to increase his WAR to tops in the NL.

But beyond WAR, which is not a proven metric on its own, Molina positively affects the game in all four aspects of baseball: Batting, Pitching, Defense, and Baserunning.  That is not arguable.  A quick look at game logs shows the Cardinals were a .500 team with Molina either not starting or hurt from the All-Star break until off the DL August 15th with a knee injury.  They finished the season 32 games over .500 meaning they were 32 games over .500 with a healthy Molina in the lineup and behind the plate.

Quote theoretical wins above replacement all you want for McCutchen, Gomez, and Goldschmidt.  The Cardinals were literally 32 ACTUAL wins better with Molina vs. his replacement in 2013, and there is no fuzzy math there.  Yadier Molina won’t have the trophy in his case, but he was without a doubt the most valuable player in 2013.

There you have it.

For more information from Chris Tyrrell, follow him on Twitter: @LuckySTLFan

This was a guest post for @stlCupofJoe‘s Sports Blog

Advertisements

4 thoughts on “The WAR against Molina (by Chris Tyrrell)

  1. Excellent, excellent article. The writers are simply daft. The greatest catcher in the game has a first-rate year at the bat. That’s it, it’s over, MVP. The writers used to understand the whole value of a great catcher: Yogi, Campy. In the early MVP voting, even light-hitting catchers got a lot of attention: Bob O’Farrell, Johnny Bassler …

Go Ahead! Leave a Reply:

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s