Ryan Hanigan Proclaims to Runners that Thou Shalt Not Steal

I don't know the religious affiliation of Ryan Hanigan, but he sure takes the 8th Commandment very seriously.  This season he has been tremendous at keeping base runners at bay, throwing out 56% of the runners who have tried to steal on him.  That rate is easily the best in the Majors for any catcher who has caught at least 25 innings this season. 

I was interested in precisely how much Hanigan has helped his cause with his arm.  A couple of people have devised methods for measuring the value of a catcher's arm, but I felt like they were a little short, especially given the data that is currently available to us.  So, I decided to take a stab at it myself.  I based some of what I did off of Chone Smith's work at BBTF back in 2005, and I don't know if he's updated it, but I felt like it could be improved upon in a spot or two.  Here are the results that I came up with.

As you can see, Hanigan has been a badass with his arm this year.  Saving 5 runs may not seem like a lot, but we're just over a third of the way into the season and Hanigan has only played in half of the innings , and yet he's still saved 5 runs over an average catcher in that same amount of playing time.  Granted, he's likely to regress as the season continues, but to this point, he's been the best throwing catcher in the Majors.  It's especially exciting to see his name above two of the greats in the game.  Not bad for a rookie.

Ramon Hernandez has also thrown out quite a few runners himself, but he's not really in Hanigan's league so far this season.  You'll also notice former Red David Ross in at #5.  In limited playing time he's having one of his best seasons ever throwing out base runners.

For those that are interested in how the sausage is made, I'll explain my methodology after the jump.

First of all, I took the obvious path of calculating the run value of each catcher's stolen bases allowed versus their caught stealings where the catcher got an assist.  This is a little different than what Smith did back in 2005, because he included all caught stealings, which means he included pick off-caught stealing, which the catcher does not get an assist on.  Like Smith, I used the linear weight run values for SB (-0.19 runs) and CS (0.44) runs to tally up those values.

The next thing I added that Smith did not have, mainly because it wasn't available back then, is pick offs by the catcher.  I gave those the same run value as a caught stealing since it is essentially the same thing.

Finally, like Smith, I thought that it was important to give the catcher credit for his reputation.  In his prime, Johnny Bench didn't throw out a lot of runners, but that was mainly because people stopped running on him.  Obviously there is some great value in that.  There are a couple of differences in the way that I calculated reputation value though.  First, Smith estimates the number of expected stolen base attempts based on catcher innings.  Thanks to Baseball-Reference, we can be a little more precise.  Now we know just how many times a catcher had to face a runner on 1st or 2nd with a base open in front of them.  Using this information, we can estimate the number of runners who choose not to attempt a steal against a catcher compared to the rate at which a base runner attempts a steal against an average catcher.  This net value will give us an indication of how comfortable runners are at attempting to steal against a specific catcher.

The second difference in the reputation value calculation that I came up with is that I don't credit the catcher with keeping a double play in tact, rather I credit him with saving the runs that an average catcher would give up in those extra opportunities where the runner does not run.  For instance, if a catcher has 400 stolen base opportunities and runners only attempt to steal 5% of the time compared to a league average of 10% (for instance), then that catcher's reputation has stifled the running game by 20 steal attempts.  If the average base stealer is successful 75% of the time, then theoretically that catcher has prevented 15 extra stolen bases as well as 5 caught stealings.  Using linear weights, this catcher's reputation saved his team 0.65 runs.

NOTE: Obviously, depressed or increased stolen base attempts aren't all the fault of the catcher.  Some catchers are lucky enough to catch multiple pitchers with excellent pick off moves, so they could be benefiting from that.  Also, there is the issue of game situations.  Catchers who frequently catch in blowouts are probably less likely to have stolen bases attempted on them.  I know this reputation metric isn't pristine, but as you can see it's also not a very big chunk of the pie either.  I feel like it's still a good way to boost up numbers for a catcher like Yadier Molina whose reputation clearly keeps him from throwing out more runners.

I totaled all of these run values up to get a net runs saved value and then compared it to the average catcher rate for that player's playing time (based on SB Opportunities).  That is how I get the Runs Saved Above Average value.

I'm interested in hearing what critiques you might have of this, so please fire away.  I'm sure I've probably recreated the work of somebody out there, but I wanted to do this my own way just for the fun of it.  If you feel like there is something that I've left out or someone who does it better, please point me to their work because I'd love to continue to refine this.