ST. LOUIS, MO - AUGUST 15: Allen Craig #21 of the St. Louis Cardinals flips the ball to the turtle covering first base against the Arizona Diamondbacks at Busch Stadium on August 15, 2012 in St. Louis, Missouri. (Photo by Dilip Vishwanat/Getty Images)
jaime garcia is scheduled to make a start for the cardinals on sunday, after a lengthy rest to allow a partial labrum tear to heal and several rehab starts that were mostly successful.
just to check in on one of our narratives from earlier in the season, let's take a look at how jaime is doing with the two scenarios allegedly causing him stress: games on the road and following up on errors.
jaime's home/road split in xFIP: 3.70/3.71. his FIP and his ERA do show a split, but mostly because he's allowed 2 HR on the road and none at home.
as to whether errors cause jaime to break down mid-game, the whole premise of the claim was that he permitted a surprising number of unearned runs in 2010 and 2011 (15 and 23). so far this year, he's allowed exactly one unearned run.
hopefully, we can start to put behind us the stories of his mental fragility.
this year, prior to his injury, jaime had a frustratingly large gap between his FIP and ERA - 2.81 and 4.48. his FIP is top-notch and his ERA is maybe better than replacement value. of course, that disparity is likely to narrow. he probably won't maintain the .357 BABIP against or the 67% LOB rate. he'll allow more home runs, too. his xFIP for the season so far is 3.70.
of course, what the projection systems and the peripherals don't know is that he's coming back from a labrum repair.
it's hard to know exactly what to expect from him. his stats overall in the minors looked fine, but it's far too small a sample size to draw any conclusions. he did give up three homers between three starts in AA and AAA and walked three in his last start. a rocky re-entry is well within the realm of the possible.
a hallmark to watch for: his velocity has been down a tick this season. where his fastball had previously averaged 89-90 mph, it sat around 88 this season. the peak velocity was down as well. in prior seasons, he turned his fastball peak velocity up to 92 or 93. this season, he didn't break 91. watch the gun readings on sunday to get a sense of where he's sitting.
i don't know that i'd throw a party if he hits 92 mph or that i'd despair if he hovers around 87-88 mph. the former could be overthrowing that's unsustainable, or that latter could be latent effects of the recovery that will wear off after a couple starts. still, i think it's a trend worth watching.
a healthy jaime garcia is clearly a better option than joe kelly. i was a little bit surprised by the number of commenters and media types who openly wondered about whether to put garcia back in the rotation at all. the only reason not to do that would be either a) an excessive trust in ERA as a predictive stat or b) a better knowledge of the specifics of garcia's recovery and the innards of his left shoulder than any of us possess.
one thing that was surprising to me was to realize how long garcia's been out. as of today, joe kelly has actually thrown slightly more innings than jaime garcia in the majors (68 v. 66 innings).
ZiPS is still remarkably pessimistic about kelly's prospects going forward, predicting a 5.10 FIP for joe kelly the rest of the way. using his peripherals directly, he has a 4.01 FIP and 4.06 xFIP which makes him an almost league average starter. his ERA has been regressing towards his FIP for some time and now sits at 3.41.
you have to give kelly credit. he has done a fantastic job on the mound, exceeding my expectations of his debut by quite a bit. he still walks more guys than a fellow who relies more on groundballs than a gaudy strikeout rate should. my subjective and flawed recollection of his starts indicates that he gets hit pretty hard with some frequency. on the other hand, his 51% GB rate has helped him out of many a jam with a timely double play. he seems like a great candidate for the back end of next spring's rotation.
still, i don't think there's anything about his game that makes me think that he's a better option than jaime as a starter. i may find myself eating my words if jaime's shoulder is not well, but i don't think that we can do anything but guess whether he's healthy on the mound on sunday.
what is immediately exciting, though, is the prospect of seeing joe eating some innings in relief. if he dials his fastball up a little more and uses his change and his slider less, he has great potential as a reliever. the typical wisdom is that an average pitcher loses about a half a run from his ERA in transitioning to the bullpen, and puts half a run on moving to the rotation. if anything, i'd expect kelly to thrive a little bit more in the bullpen than the average convert. that should make him one of our more effective relievers.
i'm also surprised by the continued mixing of lance lynn into this discussion. i think the most common proposal has been to move lynn to the bullpen and leave kelly in the rotation. i find that really strange, since it's pretty manifest that lance lynn has been an outstanding starter, with a 3.65 ERA/3.58 FIP/3.69 xFIP. i guess the idea is that a handful of bad recent starts justify the move.
mostly, the arguments in favor of the "lynn is tired" theory has been an exercise in artificial endpoints. lynn's velocity is exactly where it should be and has not declined at all. july was one of his best months overall, with a 2.67 ERA, 2.97 FIP, and 3.30 xFIP. his last two starts were terrible, sure (his only starts so far in august). but two starts hardly begin to make a trend. if every pitcher who throws two bad games back to back should be moved to the bullpen, we're not going to have much of a rotation.
and the general too-many-innings-pitched argument doesn't hold. last season, kelly threw 132 innings. in 2010, he threw 103 innings. so far this year, he's thrown 141 between AAA and MLB. last season, lynn threw about 110 innings. in 2010, he threw 164 innings. this season, he's thrown 138. both have thrown plenty of innings in their time. there's no reason to believe either is incapable of managing a starter's role through the end of the season. more importantly, neither is particularly better off in terms of total innings thrown so far this year.
i think keeping garcia and lynn in the rotation is the right step. keeping kelly in the bullpen will improve the team overall, and keep him handy to back up jaime, in the event jaime turns out to be less-than-fully recovered.