The tenth in an alphabetical and irregularly updated series of seasons-in-review for each of the players we predicted last winter. (All entries are linked in the right-hand sidebar, below the LL Exclusives.)
LL Community: 4.89 FIP (n=14)
PECOTA: 5.14
Bill James Handbook: 5.20
ZiPS: 5.21
Marcel: 4.86
Actual: 4.68 (4.79 xFIP)
Again, as was the case with Felix (and all the other pitchers), we went with FIP instead of ERA. Why? I don't know, but it seemed like a good idea at the time. I'm sure it has something to do with FIP being more predictable and less dependent on good/bad luck, but ERA is more reader-friendly, so I think we'll do that the next time around. Maybe both. Anyway.
Not a bad job by the LL community here - they were second-closest by a comfortable margin ahead of PECOTA, and barely fell behind Marcel. It wasn't just a matter of being "less wrong," either; Gil's final FIP fell well within the range of statistical error for the projection. I don't know if this was a case of optimism panning out or Mariner fans seeing something that other forecasting systems didn't, but either way LL beat the most advanced computer programs on Gil Meche. Neat.
If you remember, there was reason to be at least a little optimistic about Meche back in March. Coming off an absolutely godawful season, Gil needed to make some changes, and that's exactly what Raffy Chaves set about putting in motion when he taught Gil how to throw a two-seam fastball. The reasoning was threefold: (1) to get him more comfortable pitching in the zone, (2) to make him more efficient, and (3) to induce more groundballs. I was on board with the idea from Day 1, mostly because New Meche couldn't possibly be any worse than Familiar Meche. To put it medically: if you have a dead guy, and you hear about this experimental procedure that carries the potential of making him somewhat less dead, you do it, because you can't really go any lower. Plus, two-seamers are awesome. A good two-seam fastball is easily my second-favorite pitch, just behind a deceptive straight change, because it keeps the ball on the ground, and balls on the ground can't fly over the fence. So, yeah, it was a cool thought.
Unfortunately, good ideas require good execution, and that's where Gil was lacking to start the season. If he ever entertained the thought of using the two-seamer as his go-to pitch, he abandoned it almost instantly, going back to his preferred pitch-away-from-the-bat approach and only coming over the plate when he (A) had to, or (B) made a mistake. As a result, his strikeouts were up, but so were his walks, and after two months he looked like Familiar Meche on hyperdrive.
Then June happened. Over five starts, Gil allowed just seven runs, posting a 2.6 K/BB in the process. The streak wouldn't end there, either, as he overcame a lousy performance against the Angels to tame the Tigers, Blue Jays, and Yankees in order. Suddenly Gil's ERA stood at 3.83, and surrounded by a horde of disappointments he was looking like the ace of the staff. Rather than being due to a new two-seamer, it appeared that Gil's improvement was borne of a few minor mechanical adjustments, and he had everyone's attention. Maybe, just maybe, he was finally turning into the pitcher so many people thought he could be for so many years.
It didn't end that way. Through the final two months of the year, Gil's FIP was a boring old Meche-ian 4.89, with a Meche-ian 1.47 K/BB and a Meche-ian 1.15 HR/9. He finished the season even worse than he started it, and fans who had been swayed by his run of good pitching in the middle of the summer swore off ever believing in Gil again. He concluded the year appropriately enough, walking five batters in Oakland's AAA lineup and needing 125 pitches to get through six innings. Classic Gil. The second he got pulled, Mariner fans everywhere breathed a deep sigh of relief, armed with the knowledge that they'd almost certainly never have to watch Gil trot his sorry ass out to the mound in a Seattle uniform ever again. After more than a decade of underachieving, the Gil Meche Era was finally over.
Of course, the offseason does funny things, and with Matsuzaka out of the picture and the rest of the pitching market looking paper thin, there's a little chatter over bringing Gil back (on purpose). The idea is that, with only Felix and Washburn penciled into the 2007 rotation, we need all the starters we can get, and Meche is a supposed known quantity who wouldn't have to worry about league switches or park adjustments or what have you. Plus, there's the old "he's got great stuff" argument, that he's a guy who's only now starting to come into his own. And why would you want to let someone like that get away?
See, here are the objective problems with that off the top of my head, arranged in no particular order:
(1) Gil's looking at something around $24m/3yr.
(2) He isn't really a known quantity at all, considering the differences between his '05 and '06 seasons.
(3) There are better values out there.
(4) For every Jason Schmidt who looked like this and took off, there are a dozen Shawn Esteses who didn't.
(5) Gil Meche isn't very good.
#5, as you can imagine, is the most important. We'll do this by comparisons: in Gil's "peak season," his FIP ranked between Jeff Suppan and the disintegrated remains of Brad Radke. His strikeout rate was fine, placed between Dave Bush and Dan Haren, but around him in walks were Ted Lilly and Paul Maholm. Quite simply, his numbers don't place him in very flattering company. He's a borderline league-average pitcher who, most of the time, is a little worse than that. The fact that he's somewhat similar to Suppan and Lilly, two other guys who'll get too much money, doesn't justify the contract that Gil's going to get. If a couple people do something stupid, and a third person comes along and does the same thing, the third person isn't any smarter than the first two. Gil firmly belongs to that middle-tier of free agent arms that have historically been wildly overvalued, and these guys, more than anyone else, are the ones you need to avoid.
It's tempting to look at Gil's career-high strikeout rate and think that he's on the up-and-up, but it's not really that simple. Along with the whiffs came a ton of walks, and his K/BB was better in 2004. The only thing that kept his ERA under five was an increase in his frequency of groundballs (perhaps due to the occasional two-seamer), reducing his homer output. Of course, his GB% is still below the average as well; he's just slightly less of a flyball pitcher than he used to be.
I understand the reservations some people have with FIP, but it's the best individual tool we have right now to evaluate pitching performance in one fell swoop, so I want to use it to demonstrate to you just how much Gil still needs to improve before he's a legitimately good pitcher. Let's assume for the sake of simplicity that Gil's 2006 strikeout and home run rates are static, and that the only thing changing are his walks. The following is a quick list of FIPs adjusted to a new walk rate relative to the one he just put up:
2006 FIP: 4.68
-10% BB: 4.55
-20% BB: 4.41
-30% BB: 4.28
-40% BB: 4.14
-50% BB: 4.01
-60% BB: 3.87
Keeping everything else the same, Gil would need to trim his walk rate by 51% to post an FIP under 4. I don't have a list handy of guys who've been able to do that in the past, but I guarantee you it's pretty short, particularly for pitchers at 28+ years of age. The average pitcher development curve isn't like the one for hitters; pitchers peak earlier and decline sooner. Late bloomers aren't as common on the mound as they are at the plate. Obviously there are exceptions, since the average doesn't apply to everyone, but giving Gil the kind of money he's likely to command only makes sense if you're absolutely convinced that he'll hit a new peak in the near future. And as far as I know, there's no way of knowing that. $24m/3yr is paying Gil to be something he's never been for more than two or three starts at a time.
Gil's probably always going to be reputed as possessing terrific stuff, and because of that people will consider him more likely to break out than any number of other pitchers. And while there is some truth to that, his arsenal isn't anywhere close to being as great as so many claim it is. We've all seen him pitch enough to know that he can miss a few bats when he's going well, but what about the rest? You'd think that "great stuff" would show up in the numbers somewhere, but it doesn't, not in K% (which hasn't been consistently good), and certainly not in HR/FB% or LD%. His repertoire is fine, and easily good enough for the Majors, but it's not so good that he can coast by on stuff alone. He needs to have both control of his pitches and a solid approach to succeed, and so far in his career he's shown neither on a consistent basis.
There's just no good reason to make this commitment. While Gil's ratios have changed over the years, his player profile hasn't - he's still the same old #4/#5 starter with decent stuff and a stupid attack. He has the potential to improve, but we've been saying that for more years than I can count, and at some point you have to give up and let him try to tap into his ability somewhere else. It's essentially a relationship where Gil and the Mariners fight all the time about the same thing. Eventually you have to realize that neither party's going to change, at least not in the current situation, so they have to break it off and go their separate ways, smiling and waving when they see each other but never trying to rekindle an old flame. Via free agency, Gil Meche and the Mariners have been granted a divorce. It's time to move on.
Gil might turn into Jason Schmidt or Chris Carpenter. He's got a better shot of doing that than, say, Joel Pineiro or Jake Woods. But evaluating players and building a roster based on their best possible outcomes is a great way to get yourself in a heap of trouble. I've been saying for a while that Gil seems like a great candidate to go all Esteban Loaiza on the rest of the league, but is that really worth protecting? And considering how much help he's gotten from Safeco over the years (146-point ERA difference between home and away), are the odds of Gil taking the leap even that good to begin with?
It's time to end this thing. Offer Meche arbitration, let him walk, and take the draft pick, happy that, after 11 years of one-sided service, we finally have something to show for that bust of a first-rounder. Gil, your time in Seattle is up. Now go away.