Bob Smizik's latest piece, which suggests five changes the Pirates might make, is actually pretty interesting.
Four of them (getting Freddy Sanchez into the lineup, getting Oliver Perez out of the rotation if he can't be fixed, moving Chris Duffy and Nate McLouth lower in the order unless they start hitting, and removing Humberto Cota as Perez' personal catcher) aren't revelatory. I agree with all of them except the one about what to do with Perez, and only because Dejan Kovacevic clouds the issue in today's Q+A, in which he says Perez can't be sent to Indianapolis without passing through waivers. (Since he says Perez has an option, this doesn't sound right - can anyone clarify this?)
But Smizik's other suggestion - to use Mike Gonzalez in the first inning rather than in the last - is, at least, provocative. But in the end, it's unworkable. Pirates' starters have been horrible in the first inning, it's true. But there is no particular reason to think that the Bucs' starters will continue to be this bad in the first inning as compared to, say, the third. And if there is some reason I'm not aware of (like, say, "the jitters"), I don't know why that wouldn't continue to be a problem in the second inning if someone else pitched the first.
The reason a team saves its best relievers for the late innings is because there's a greater understanding then of what's at stake. If your team is down 12 runs in the eighth inning, you know the chances of your team coming back and winning are very slim, so it's safe to throw Ryan Vogelsong out there and let him take his lumps. But if you're up one run, it matters a great deal how your team pitches that inning; the outcome of the game very well might depend on it. So if you're the manager, you want to put one of your best relievers out there.
In the first inning, a team has no idea what is at stake. If Gonzalez comes in and pitches the first inning and then the starter comes in and gets hammered in the second, that first inning by Gonzalez has been wasted on a game in which his contributions have largely been rendered irrelevant.
The problem might be better explained by this fruitless thought experiment. Gonzalez can only pitch 70 or so innings a year. So if you want to use him as a first-inning guy, which games do you have him pitch? You don't know, because in the first inning you have no idea how a game will turn out. And if the Bucs do use Gonzalez and, say, Damaso Marte and Roberto Hernandez to pitch first innings, then they'll only have the back-end relievers to pitch the eighth and ninth innings of close games. The '06 Bucs actually have a pretty good bullpen, but the result of using the good relievers in the first inning could be that the problems of the first inning are effectively shifted to the end of the game, as worse relievers are forced to take on more high-leverage situations and give up more runs.
Ultimately, using Gonzalez as a "starter" won't work. It's a fun idea, though.