Comparing catcher defense: Johjima versus Rivera
I’ve been trying to ignore a Bob Finnigan column since Sunday (M’s notes: Johjima’s defense lags), but it’s not working. So I’m going to take a short break from the book rewrites and tear into this.
Kenji Johjima has done a solid job offensively in his first season of U.S. baseball, and will rate consideration for Rookie of the Year as a result.
But according to the statistics, he has not settled in as well defensively.
(breathlessly) What statistics are those, Bob?
In fact, his defense is considerably less than that of backup catcher Rene Rivera, whose offense isn’t close to Johjima’s.
The biggest disparity is in their respective catching earned-run averages — the ERA of the pitchers they catch.
Oh. Is that it? Really. I don’t expect Finnigan to do serious baseball statistical analysis, but to hang your hat on that is still disappointing. And Johjima’s defense hasn’t as amazing as we’d hoped, so I don’t expect Finnigan to shower with him praise. But this is ridiculous garbage. You might as well quote their respective MPG.
In addition, Rivera had thrown out 31.3 percent (5 of 11) of base stealers, while Johjima has thrown out 26.3 percent. (15 of 42).
Uh huh. Here’s the thing: that’s wrong. Since Saturday’s game: Saturday 1 SB (Moyer/Johjima), Sunday 2 SB (Hernandez,Green/Johjima), no SB yesterday
So when Finnigan wrote this, Johjima was at 42 SB and 21 CS. So of the 63 attempts, 21 were caught = 33% caught stealing.
It looks like he just read the stat lines as SB = attempts, CS = caught stealing, and then still managed to read the stats (or was working off… I’m not sure what he would have had to work off to get those numbers) and do the math both wrong. That takes some doing.
Plus, that’s a ridiculously small sample size.
But back to his basic argument on catcher ERA.
Yet breaking down the ERA by individual pitcher, some of the differences between Johjima and Rivera are stark:
Felix Hernandez: Johjima 4.73, Rivera 2.94; Gil Meche: Johjima 5.21, Rivera 2.17; Jamie Moyer (into Saturday night): Johjima 4.51, Rivera 3.16; J.J. Putz: Johjima 2.88, Rivera 0.00; Jarrod Washburn, Johjima 4.57, Rivera 1.32.
This would be a good way to compare the two catchers if that statistic meant anything, but it doesn’t. It’s the Game-Winning RBI of defensive statistics. Especially given the small sample sizes involved – Rivera catches a guy at Safeco, he’s going to look better than Johjima catching the same dude at Colorado. Glancing over the game logs, I see he had one start at Boston, and that’s pretty much it for severe hitters parks. I’d run the numbers but there’s no point.
There’s no point because catcher ERA is essentially meaningless. Seriously. All that stuff about calling a game and whatnot? Barely matters. It’s something to fill the three hours of broadcast time and, in this case, a couple of column inches. Not only is ERA not a particularly good stat to measure pitcher performance with, but its problems are magnified when you attempt to slice it and assign credit or blame based on the results.
Keith Woolner did a study back in 1999 (“Field General or Backstop?“) and followed that up with a fairly mammoth simulation-based work in 2002.
I’ll summarize the argument over catcher ERA:
Pro-catcher-game-calling: a catcher’s effect on the game is masked because there’s so much noise
Skeptics: okay, but that means it can’t be that huge
It’s a lot like the clutch hitting argument. There are clutch hits, but the traditionally defined clutch hitter – someone who performs better than expected in whatever you define clutch situations as – isn’t a common species of hitter. If there are hitters who have some ability to do better, that ability is lost in the noise.
Woolner’s conclusion is well-stated:
[T]here is very little practical difference between saying there is no ability, and that there is an ability that can’t be reliably detected. Knowing that there’s no way that CERA or similar measures can indicate ability, and that any other test or scouting report can’t be validated against actual results, means that there’s no actionable knowledge to be gained.
Catcher ERA is interesting, but you should regard it in the same way you look at the random stats they display on the scoreboard (“Mike Morse is hitting .417 in home games this year”).
Now, is Rivera a better defensive catcher than Johjima? Sure, why not. But putting Rivera behind the plate doesn’t turn a bad pitcher good. Any catcher who could reliably shave a run off the score every time they got a start would be the best defensive catcher in the history of baseball and would make a kajillion dollars. That catcher doesn’t exist, and never has.
Update: It’s Finnigan, folks. Finn-i-gan. Don’t crack on him for doing crappy research and/or math if you can’t spell his name right.
Comments
54 Responses to “Comparing catcher defense: Johjima versus Rivera”
Leave a Reply
You must be logged in to post a comment.
and I’m not sure we will be able to judge his game calling for a while yet– we don’t know how much he’s been told what to call this season, a season spent learning the pitchers (his and the rest of the league — not to mention the language, the culture, hitting etc etc), we don’t know what will happen if he has a different manager (pleaseohplease)
oooh, Gas just asked Krueger what he thinks of catcher ERA– Bill sez well, if you had a significant sample of innings on the same team (and he points out that Rene hasn’t) and if there was a significant difference in ERA it would be something discussable … Nice wiggle out of it, Bill 🙂
His evaluation? Both catchers are light on experience, with Rene a little stronger defensively but Kenji a better hitter.
If Rene was that big of a difference maker in the ERA department the club would have caught on by now. If Rene could actually turn Washburn from a 4.57 ERA pitcher into a 1.32 ERA pitcher then he would be worth his weight in gold.
Gilligan…I mean Finnigan is just pulling stuff out of his nether-regions without “wasting: his tome doing actual research.
I see we have another night of Jones getting big league experience on the bench while Willie gets to spread his *grit* all over Centerfield.
Damn beer….”wasting” his time. I have no idea what he does with his tome.
Dave, while your “weeded out” assertion is generally true, I would counter that Johjima’s supposedly distinctive approach to game-calling made it to the bigs without the typical minor-league weeding out. If all catchers make it to the bigs with a by-the-book approach to game calling we would naturally expect to find little if any discernable (or statistically significant) difference in CERA. We can legitimately hypothesize that Joh’s game-calling is different than Rivera’s.
A true statistical test would be a linear regression of each pitching appearance using the following model:
runs = a(catcher) + b(pitcher’s RA) + c(opp. RA) + d(IP) + e(park factor)
where for (catcher) Johjima=1 and Rivera=0.