A tale of two Gils, indeed. It was the best of times, it was the blurst of times, where "blurst" is probably the best adjective we've got to describe Meche's eye-gougingly bad first half. To continue with the extended Simpsons reference, the April/May Gil Meche was about as good a pitcher as a bunch of monkeys with typewriters are good journalists. Many of us were ready to write off 2004 as a lost season, perhaps the fallout of 2003 fatigue.
And then something funny happened.
Meche got demoted to Tacoma, where he essentially kept on pitching just as poorly as he had in Seattle - a 1.67 K/BB, a 5.05 ERA, and five more unearned runs that pushed his RA up near six. Nevertheless, the Mariners decided to recall him in late July, and he was a different pitcher. Gone were the control problems, the constant nibbling, and the corresponding runs. Before we knew it, Gil became our second-best pitcher down the stretch, spinning a 3.95 ERA and spreading hope that a complete game shutout of Boston on September 12th might be enough to turn his entire career around.
So, it's analysis time.
ERA: 5.01
RA: 5.15
dERA: 4.81
cERA: 5.01
eERA: 5.20
BABIP: .300
Of course, that doesn't really tell us anything; the second half Meche was so different from the one that started the season that we need to treat the two Gils as separate entities.
The difference is stunning. Meche cut his walks by nearly two-thirds, lost a sixth of his strikeouts, and saw his home runs increase at the expense of some doubles and triples. He also became a more extreme flyball pitcher.
Check out that last column, though. For an idea of how significant that difference is, a .377 BABIP over a full season would have been the worst in baseball by 39 points. A .262 BABIP, on the other hand, would have ranked as the ninth-best in the league.
I think we need a new table.
Looking at the last two columns, we can see that Meche allowed roughly as many runs as you'd expect in each half, given his walks, extra-base hits allowed, and batters faced. Thus, the wildly different ERAs aren't the result of, say, Meche pitching inordinately well with runners on base in the second half.
Now look at the dERA column. If you aren't familiar with this header, it refers to a pitcher's defense-independent ERA - that is, the ERA you could expect him to put up in a league-average defensive context. A standard BABIP is somewhere in the neighborhood of .295, well below Meche's figure in the first half and significantly higher than his second half split.
With this in mind, we can infer two simple points:
- Gil Meche was unlucky in the first half
- Gil Meche was lucky in the second half
No matter how poorly he was pitching over his first ten starts, no matter how many times he floated a fastball over the middle or spun a hanging curveball, there's no getting around the fact that a .377 BABIP is just incredibly bad fortune.
Likewise, no matter how well Meche was pitching down the stretch, no matter how many times he put a breaking ball on the outside corner, there's no getting around the fact that a .262 BABIP is incredibly good fortune.
Of course, we can account for his second half much better than we can his first; Meche became more of a flyball pitcher as the summer wore on, and these types of pitchers tend to allow fewer hits per ball in play than a standard arm or a groundballer. Throw in the fact that Meche was pitching in front of a good defense in a beneficial environment, and you've got all the necessary ingredients for a low hit rate. Not .262 low, mind you, but lower than usual.
So what happened in the first half? There was obviously some funny stuff going on in the field behind him, but this doesn't completely absolve Meche of blame; his line drive percentage was higher than the league average, which will lead to more balls falling in for hits, and his generally poor command led to a bunch of missed spots, which gave batters easier pitches to hit. The point is that there is a difference between the pre-ASB and the post-ASB Gil Meche that goes beyond defensive performance behind him.
When he came back from Tacoma, Meche was throwing strike after strike after strike. It's been suggested that he had a lengthy chat with some of the coaching staff in the minors after which he changed his approach on the hill, but I think this is just an example of people trying to find a reason for Meche's turnaround. So for whatever reason, he came back to the Majors and pounded the strike zone, slashing his walks at the expense of a few strikeouts and forcing hitters to do the work, rather than handing them free passes to first. The approach worked, as he tossed seven quality starts in the second half, compared to two in the first. He left the ball up in the zone a little bit, which led to an increased rate of fly balls and a few extra homers, but he actually lowered his extra-base hit rate. Some of this is the defense, and some of this is Meche. With his home run and 2b/3b rates in mind, you could even suggest that, like Jamie Moyer, Meche got a little unlucky in the second half in terms of balls leaving the park. This could be true, but certainly not to the extent that Moyer got hurt.
Meche v2.0 was helped by his defense, but he was obviously a better pitcher than the one that started the season. So, there are two questions:
(1) Are Gil's improvements for real?
(2) If so, would this allow him to have continued success?
As for #1, the only way to know for sure is to see how he does in 2005. However, his K/BB in the second half was different enough from his track record to stir some skepticism. Never before had he put up such a low walk rate, and his previous career-best K/BB was 2.67 in Wisconsin in 1998. I think that Meche has improved as a pitcher, but I expect him to give back a little of that K/BB going forward, to be more in line with something in the low 2's.
As for #2, yes, you can have success as a strike-thrower with a home run problem. Mark Buehrle and Zack Greinke's 2004 performances were eerily similar to Meche's, and they both posted ERA's below 4. Buehrle allowed more singles and fewer doubles/triples, as a more neutral pitcher, so Greinke is the better comparison. He also had a low (.266) BABIP to go with a 3.97 ERA, suggesting that his real ability level for 2004 was an ERA somewhere in the low 4's.
That's what I expect to see from Meche next year. If he can avoid giving back too much of his improvement, then he should be a reasonably effective arm who finishes the year around 4.30 in 30+ starts. He won't be the ace that a lot people think he'll be, but barring injury, he should have the best full season of his career, setting up for an intriguing few seasons as he enters his physical prime.