FanGraphs Baseball

Comments

RSS feed for comments on this post.

  1. You should probably try to remove any data that involves a knuckle ball (i.e Kevin Cash / Tim Wakefield).

    Comment by Scappy — February 25, 2009 @ 11:46 am

  2. The point of the process is to first look at which pitcher each catcher is working with, and develop an expected value. So instead of tossing them out, they are accounted for.

    Despite Charlie Hough once throwing 65 PB in a single season, Wakefield is the career leader in PB% at .045 (10x the MLB avg) with Hoyt Wilhelm at .044 and Hough at .043. Then drops down to Barney Schultz, Jared Fernandez, Phil Niekro, Tom Candiotti…all the knucklers.

    Rounded to nearest .001, with MLB avg .004, Cash in 2005, based on his pitchers was expected to have .004, had .005 (slightly above avg), rating is .006. In 2008, expected was .014 (a lot of Wakefield), observed was .014, the same, so rating is .004, league average.

    Paul Mirabelli, from 2002 to 2007, expected .011, .017, .018, .019, .011, .017, observed .011, .014, .015, .009, .011, .008, so never higher than expected, sometimes much lower – so he was better thn expected at avoided PBs

    Comment by Brian Cartwright — February 25, 2009 @ 12:01 pm

  3. I love looking at how historic players rank against current players. I find it very interesting to see someone like Yogi Berra pop up ahead of current stars even thought the dynamics of athletics has changed so greatly.

    – Jeff

    Comment by Jeff — February 25, 2009 @ 1:34 pm

  4. This is one area that probably hasn’t changed that much, the pitcher and catcher playing catch…which pitchers are hard to catch and which catchers excel and and which struggle.

    But also, with the list of active players, to get an idea of what the range of ability there is on defense for catchers, and to think that Gary Carter once had a +28 season.

    Comment by Brian Cartwright — February 25, 2009 @ 5:59 pm

  5. I just don’t see how Ramon Hernandez can have anything positive related to his defensive skills. Care to elaborate; because he was slow, lazy and notoriously bad at blocking balls in the dirt last year.

    Comment by Mike — February 26, 2009 @ 12:58 am

  6. In 8 of Hernandez’s 10 seasons in MLB, his ratio of observed WP+PB to expected has been between .72 and .77, in 2000 was .97 and 2006 1.22. The MLB avg for WP% is .016. In every year thru 2006 his expected was .015 or .016, almost exactly average. In 2007 his expected was .021, in 2008 .025, as there were pitchers on the Orioles those years with a history of throwing an above average amount of wild pitches with catchers other than Hernandez. His observed pct increased to .014 and .019, but stayed in the exact same ratio to expected that he had in 6 of his first 8 seasons.

    Last year there were 46 WP and 10 PB with Hernandez catching. That number of WP was 2nd highest in 2008 behind Bengie Molina’s 54, but my formula expected him to allow 63 WP and 11 PB, the 63 expected WP the highest in MLB in 2008. Of the Orioles top 7 in IP in 2008, only Garret Olson was avg at .016, then Burres .018, Guthrie .020, Cormier .021, Cabrera .031, Liz .033, Sarfate .033

    Comment by Brian Cartwright — February 26, 2009 @ 1:41 am

  7. Considering that Joe Mauer’s backup is apperently one of the best of all time, and that most of the Twins starting staff has only had those two catch them, is it possible he is being undervalued in this regard?

    Comment by lookatthosetwins — February 26, 2009 @ 2:10 am

  8. Just curious, how do you calculate the expected passed balls? Do you just use the career averages of the pitcher’s they’ve caught?

    Comment by lookatthosetwins — February 26, 2009 @ 2:12 am

  9. Yes, I used the career averages of the pitchers, minus what those pitchers did with that catcher. I also ran the pitcher numbers on a yearly basis, and compared it to the results using career numbers. In the vast majority of cases there was no meaningful difference in the numbers between using the pitchers yearly or career. I decided to go with the pitcher’s career to avoid as much as possible the Mauer/Redmond situation you cite, where a player would be mostly compared to his backup.

    On the other hand, by using career you run a risk of overvaluing a guy like Ramon Hernandez, if his staff does better than their career numbers – but that seems unlikely that a dozen or more players would have a random variation biased towards one direction, other than just about the mean. I think that’s vindicated by Hernandez’s 2007 & 2008 ratings, despite a high expected value, fell right in line with his previous years and career totals.

    Comment by Brian Cartwright — February 26, 2009 @ 2:55 am

  10. “The pitchers he caught that year would have been expected to throw 62 wild pitches and 20 passed balls in Freehan’s playing time,”

    Where do you get the data on how many pitches Freehan caught from each pitcher with a batter on base? It seems to me there are a lot of assumptions here about things evening out that are not only unproven, but unlikely.

    Is it just a coincidence the Uecker caught knuckleballer Phil Niekro in 1967? In fact, wasn’t Atlanta looking for a guy who could catch a knuckleballer when the traded for him?

    “that seems unlikely that a dozen or more players would have a random variation biased towards one direction,”

    I don’t see why that is unlikely. In fact, it seems to me in the sample of all major league catchers you would expect to find several examples of where that happened.

    But there is another question here. HOW did the catcher avoid passed balls and wild pitches. Was it because pitchers didn’t throw in the dirt or because the catcher didn’t call for a knuckleball.

    Comment by TT — December 16, 2009 @ 6:35 pm

  11. I do not know each pitch, but I do know each batter.

    Using Retrosheet, I created a database table that showed for every combination of catcher and pitcher, how many plate appearances there were with runners on base, and how many times was there a wild pitch or a passed ball. Then repeat by catchers only, and by pitchers only.

    So for Uecker, I know how many PAs with runners on base, how many WP and how many PB when he was catching Niekro, and when he was catching any other pitcher. If I take Niekro’s pitcher’s totals, and subtract off what he and Uecker did together, then I know how Niekro did with every other catcher. Therefor, I can compare Uecker catching Niekro to everyone else who caught Niekro. Repeat for each catcher and each pitcher, and sum the results.

    I don’t know how the catcher avoided more wild pitches, but I can measure how many.

    Dave Allen recently wrote an article at Baseball Analysts http://baseballanalysts.com/archives/2009/11/a_pitchfx_look.php that examined catcher’s ability to block pitches based on the location of the pitch.

    Comment by Brian Cartwright — December 16, 2009 @ 7:03 pm

  12. Hey there, You have done an incredible job. I’ll certainly digg it and in my view recommend to my friends. I’m sure they’ll be benefited from this site.

    Comment by rug — November 15, 2011 @ 10:21 am

Leave a comment

Line and paragraph breaks automatic, e-mail address never displayed, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>


Current day month ye@r *

Close this window.

0.081 Powered by WordPress