Mike Napoli vs. Jeff Mathis, or: Why Catchers' RAR Could be Renamed "RAJM"

I've been asking myself why Mike Napoli hasn't locked down the starting catcher gig for Los Angeheim yet for about two years now. Currently, Napoli splits time with Jeff Mathis, a former first-rounder who gets the nod from Scioscia more regularly than Napoli due to Mathis's superior defensive skills. This seems logical absent any sort of real analysis; after all, catcher is a tremendously important defensive position (probably the most important), and if a team starts hemorhagging runs behind the dish, bad things will happen.

The trouble for the Halos is that the numbers that should be driving Scioscia's decision just don't add up.

Last year, Mike Napoli hit .273/.374/.586 with 20 HR and 133 TB in only 78 games. He accounted for 17 RAR (runs above replacement) at the plate as a #2, platoon catcher!

Mathis, meanwhile, barely hit his weight, posting a .194/.275/.318 line with 9 HR and 90 TB in 94 games, all of which was "good" for a batting RAR of -18.0 (yes, that's negative eighteen).

There is no platoon advantage, either. Both Napoli and Mathis hit from the right side, and both had roughly identical splits (they hit lefties better, albeit at roughly a league average rate for righties). So not only is neither a platoon candidate in general, but even if their advantages against lefties were pushed to comically absurd levels, the fact that both are stronger against LHPs would make them poor platoon partners.

So at first glance, the winningest team in the Majors last year trotted out a catcher who couldn't stay above the Mendoza line, costing his team 18 runs versus the average farmhand, in roughly two thirds of their 2008 contests. Now, a lot of people (both Angels organization insiders and writers/bloggers) have been claiming that Mathis has improved his hitting (is it even possible to get worse than that '08 line?), and the early results this year, though beyond insignificant statistically, have got to be encouraging for the Mathis camp: in his first three games, Mathis is 4-for-10 with a double, three runs scored, two RBI, and a walk. Of course, the argument for Mathis over Napoli is that Mathis makes up for his shortcomings at the plate with what he does behind it, so if he is in fact a good enough defender to make up that 33 run defecit (Napoli loses two runs for his baserunning according to CHONE, which is the only reason the defecit isn't 35) then any added offensive production in 2009 would be gravy. But is Mathis's defense good enough to make up such a huge gap in run production through run prevention?

Not even close.

Mathis is a very solid defensive catcher, good for 9 RAR behind the dish. But Napoli isn't terrible on defense; he's on the wrong side of zero, but a -1 RAR looks pretty manageable next to all those extra batting runs. Even taking Mathis's extra playing time into account, his total RAR (hitting, fielding, and the whole nine) comes to just two. Napoli's RAR? 23. Napoli was worth 2.3 wins on the short end of a catching platoon; he's a 4-5 win player waiting to happen. Mathis was worth only 0.2, making him a replacement-level catcher in 2008.

The trouble with Scioscia's logic isn't that defense is unimportant for a catcher. Obviously, it's tremendously important (though most sabermetric analysis still rates defense at about half the importance of offense). The trouble is that Scioscia doesn't seem to be paying attention to how much better Mathis and Napoli are than one another at each aspect of the game. Mathis is clearly the superior defensive catcher, though not by an awful lot when that superiority is translated into baseball currency (runs). Napoli is clearly the superior offensive catcher, and by a hell of a lot. A 33 run advantage with the bat is absolutely monstrous. Mathis couldn't overcome it with his 9 RAR catching last year. Neither could Yadier Molina and his 13 FRAR. Nor could Johnny Bench in any of his 17 seasons (not that he'd have needed to; in his prime, the man could hit). Give Mathis Pudge Rodriguez's (arguably the best defensive catcher of all time) 1996 catching skills (33 RAR) and he just barely pulls even with Mike Napoli.

All this should put things in pretty stark perspective: for a hitter like Jeff Mathis to be worth starting over a hitter like Mike Napoli, Mathis would have to be the best defensive catcher of all time. Or he'd have to be really damned good and learn how to hit a little.

There are, of course, extenuating circumstances. As I've already mentioned, Mathis has reportedly improved his hitting significantly (though as encouraging as Mathis's first three games have been, let's not forget that in Napoli's first two he has gone 4-for-6 with 2 2B, 2 HR, 2 BB, 2 R, and 3 RBI for a hilarious week 1 OPS+ of 584). Napoli has also been injury prone, and had arthroscopic surgery on his right shoulder last October (though the procedure was described as "a cleanup"). What effects, for better or worse, this surgery will have on Napoli's offensive and defensive numbers remains to be seen, but even assuming that Napoli regresses and Mathis improves, it seems unlikely that last year's 2 RAR player (in 16 more games!) will overtake last year's 23 RAR player.

Mike Napoli may never have the durability of a Jason Kendall or the defensive skillset of a Pudge Rodriguez (or even a Jeff Mathis, for that matter!), but 120 games of Mike Napoli for an AL West team that needs to fight off the new-and-improved Oakland A's would be tremendous. At the very least, Scioscia should start playing his more valuable player more often. And although he certainly shouldn't ignore the benefits of run prevention, he needs to quantify that run prevention rather than using an overly simplistic "defense > offense" equation when deciding who should gear up more often, because right now he's essentially starting Damian Miller over Mike Piazza in two thirds of the Angels' games.