Question: In the Wright piece, was he normalizing to equivalent innings between catchers or big inning differences like between Ruiz and Schnider?

When you’re comparing pitcher/catcher duples, you get a lot of small sample samples, with all the variability that you would expect. For example, in 2010, Paul Hoover caught two of Roy Halladay’s starts. In 9 innings, he/they gave up 7 ER, with peripherals to match. If, when comparing Hoover to Ruiz, and we normalize up to Ruiz’s 202.7 innings catching Halladay, we’re saying that Hoover had over 200 innings of real awfulness, which is something that didn’t happen and the rest of Hoover’s record doesn’t support. If we don’t normalize at all, then we’re back to catcher ERA and whoever catches a higher percentage of Joe Blanton innings loses. Ruiz catches a relatively smaller percentage of Blanton innings than Halladay, Oswalt, Lee, or Hamels innings, which may help explain the catcher K/W rates you cited.

I prefer ERA as a descriptive stat and FIP as a predictive stat. I know that ERA ain’t so good for evaluating relievers, but catchers rarely get pulled in the middle of innings, so inherited runners isn’t really an issue for them. RA might have been better than ERA, but ERA works as a check and there was a football game on while I was doing this and I was occasionally distracted.

.271/.306/.394 in 424 AB

Overprojected on this list? Wieters and Santana

Underprojected? Avila and Napoli

I guess that was kinda my point about catcher defense- it’s so hard to measure right now that the results are conflicting. I’m guessing that the Wright article is not floating around on the internet somewhere? If not I’ll have to go find a copy to read somehow.

I used the methodology from a 2009 Hardball Times Annual essay by Craig Wright on Mike Piazza’s defense and why it was underrated. When I started, I expected that Ruiz would be better than others by, say, a .15 ERA difference, because he has a good reputation and the difference between Mike Napoli and Jose Molina, Jeff Mathis, and others in Anaheim was about .25 in ERA. I was shocked by what I did find: Virtually all indicators bad for Ruiz.

I repeated the process for the 2010 season and the results were Sardinha (sample size alert), Ruiz, Schneider with peripherals and Paul Hoover’s results varying. This is more what I expected to find.

Maybe when the list of DH’s comes out.

Instead of looking at counting stats, why not look at the ratios:
K/BB: Ruiz: 3.39, Schnieder 2.71 (This is despite Ruiz catching over 3 times the number of innings.)

Also a serious question, is normalizing to the lower number of innings the right thing to do? It seems to me that evaluating catcher defense is really hard and not as simple as your normalization. For example, see:

http://www.beyondtheboxscore.com/2011/11/1/2497088/2011-beyond-the-box-score-catcher-defense-rankings

http://www.hardballtimes.com/main/article/another-one-bites-the-dust/

http://www.baseballprospectus.com/article.php?articleid=15093

Not using catcher ERA. Comparing pitcher/catcher duples, and then normalizing to the lower number of innings when comparing the catchers. In virtually every case, ERAs, HRs, BBs, and Ks were worse with Ruiz. Ruiz had a slightly higher K rate than Sardinha. I did the same thing comparing Mike Napoli to Jeff Mathis, Bobby Wilson and los dos Molinas in Anaheim a few years ago. The effect was less than half as large.

There aren’t any pitchers who say, “The catcher on my team isn’t the best. I like the guy I used to pitch to better.”