Brett Myers was a reliever once, back in 2007. The Phillies plugged him in as their closer during the season, and in his 53 innings as a reliever, Myers held opponents to a .229/.294/.318 line with 10.8 strikeouts per nine. The experiment was short-lived, though, as Myers was right back in the 2008 rotation after the Phillies traded for Astros' closer Brad Lidge.
Other than that, Myers has been an effective -- though somewhat inconsistent -- starting pitcher throughout his career. In 1,623 career innings, he owns a 100 ERA+. He's only a mite above average in strikeouts for his career, but he leans a bit groundball as well. He has thrown 439 innings of slightly above-average baseball for the Astros as a starter over the last two seasons, but now they aim to make him a closer to replace the departed Mark Melancon.
It's an odd choice at first blush. The Astros aren't going to be very good in 2012, and an effective closer seems like an odd priority to have given their needs elsewhere. There's more to it than that, though: the rotation is the one area in which the Astros are not lacking, and Myers might have trade value as a closer, especially since he has a vesting option that could put him under team control for 2013.
Houston's rotation, sans Myers, goes more than five deep. Wandy Rodriguez is at the top as usual. The 27-year-old Bud Norris looks like he might have turned a corner last year, and could be an average or better hurler from here on out. J.A. Happ struggled in 2011, but was an effective hurler in the 289 innings prior to that -- he still grades out as average in his career, even with his 5.35 ERA last season. Livan Hernandez is just filler, but he's not what's going to keep the Astros from a playoff spot. If they're looking for 175 innings of replacement level pitching, then Hernandez will give it to them. Zach Duke exists for much the same purpose, albeit with less of a track record of competence.
After that, there is 21-year-old Jordan Lyles. His ceiling isn't high, but scouts believe him to be a capable big league starter. Kyle Weiland, acquired in the deal that sent Melancon to Boston, is another candidate to start: like Lyles, figuring out what he is capable of in a big league rotation is likely more important than throwing Myers out there again for 30 starts. Should those two fail, or Hernandez prove to finally be out of gas, Aneury Rodriguez is available to take another spin in the rotation during his age-24 campaign. His presence also means the Astros can keep Lyles and Weiland at Triple-A for a while longer, if they so desire. The Opening Day roster doesn't need to be the one on the field for the majority of the season.
They aren't the sexiest options, but they are young, and all have the capacity to turn into something the Astros can use in the future. Myers won't be a member of the next talented Astros team whether he closes or starts, so this is simply an allocation of resources for a team that needs to know what it has in its younger assets.
How will Myers do in the role, though? His 2007 campaign featured his best fastball in terms of velocity; no surprise there, given he was throwing fewer innings and could gas it. Harry Pavlidis sees Myers sitting around 92 with the ability to touch 94 out of the pen. That's a boost from the last three years, when he averaged 89 miles per hour on his four-seamer. As his slider and curveball are average offerings in terms of swing-and-miss ability, the added velocity boost on his fastball will be good for Myers.
Myers seems odd as someone you'd switch to the bullpen -- he seems more the kind of arm you would move from the pen to the rotation. He features four pitches that he uses between 15 and 23 percent of the time. None of them are stellar, but he can throw them all for strikes, and they all miss a decent number of bats. He's capable of shouldering a heavy load, too: he averaged 196 innings per year from 2003 through 2006, and, even counting the time he missed in 2009, has averaged 175 frames per year since 2008.
You have to be an excellent reliever to justify staying in the pen: Silver found that a closer with an ERA of 2.00 has about as much value as a 3.69 ERA starter. While those are based on 2006 numbers, when offense was higher than it is now, the point remains the same. You need to be a ridiculous bullpen arm in order to justify staying in that role over starting. Myers was great in the role originally, but he wasn't Mariano Rivera, and he only did it for part of one season.
Don't take this the wrong way, though. He'll be good in his new role. Tom Tango's "Rule of 17" says a pitcher can expect to see a difference of 17 points of BABIP, 17 percent change in K/PA, and 17 percent change in homers per balls in play between relief and starting. This is a rough sketch, but using 2011 as a base, you could expect Myers to post a .278 BABIP, strike out 20.4 percent of the batters he faces, and give up homers on 3.7 percent of balls in play as a reliever in 2012. The league averages in 2011 were .291, 18.4, and 3.4 percent respectively, so Myers would be doing well by this quick-and-dirty method.
Myers in the bullpen seems like a waste given his track record as a starter, but the Astros have other arms to try out in the rotation. He might end up anchoring their bullpen, or be a valuable trade chip given his $10 million vesting option in 2013. It's definitely not the traditional thing to do with an arm like Myers -- in this age of relievers becoming starters, it's actually backwards -- but it might be the right thing for this particular Astros team.