1) Keep him in his current role.
2) Move him to closer.
3) Move him to the starting rotation.
4) Trade him
To me #1 makes the LEAST sense, although it is the easiest route. As a pending free agent he will soon be a pretty expensive set-up man, a luxury that I don’t see the Cubs splurging on in the next couple of years.
As a free agent reliever, he will be on the market as a closer, albeit a largely untested closer. If Theo and Jed really do believe that he is the best lefty reliever in baseball, then put that to the test and see if he can handle the closing responsibilities this year. Of course that means moving Marmol, which, although I like the idea of in general, there might be some argument that he’d best moved later when he, hopefully, has built his value back up a bit. Furthermore, if there is one area the Cubs seem to have good organizational depth it is with potential closers.
The option that most intrigues me is moving him to the rotation. It is nice to dream of a CJ Wilson type of transition for him. There are no indications that this is being pursued though, and I’d think that Marshall would need/want to know soon with regards to his off season training.
This of course leaves the trade possibility. Certainly possible. That said, it will be very interesting to see how the trade market for middle relievers develops. Yes, an in season trade will not bring back draft pick compensation. However, holes in the bullpen are the types of things that many teams go into the season hoping (justifiably) that young internal guys can handle the job, then, in July when the team is desperate and sees the weakness they spend on at the deadline (as both the Cardinals and Rangers did this season in a big way.)
So, would Marshall make a successful jump to the rotation? I compared him to CJ Wilson to get a better idea. Yes, there are good reasons to not simply compare two players (can you say bad statistics!) but, what the heck.
1) First and foremost, it is certainly arguable that Marshall is a better reliever than CJ Wilson was. In Wilson's best and last year as a reliever he was valued as a 2.0 WAR pitcher, Marshall has exceeded that each of the last two years with WAR values of 2.2 and 2.8.
2) Wilson relied heavily on his fastball as a reliever, throwing more than 73% fastballs (including 2-seamers). This is notable since his average fastball velocity dropped from about 93mph to 90mph with his move to the rotation. Because Marshall is less dependent on his velocity, any vulnerability due to decreased FB velocity is less than what Wilson faced.
3) Wilson threw his slider 18% of the time as a closer (along with his 73% fastballs), making him a pretty classic fastball/slider power reliever. Marshall, on the other hand threw his fastball (~30%), slider (~40%), and curve (~30%) almost equally in 2011, interestingly dropping his cutter which had made up about 10% of his pitches in 2010 and 20% in 2009.So, Marshall is arguably a better pitcher, with more good pitches in his arsenal, who is less dependent on his velocity. Furthermore, he came up as a starter so he has a history of throwing more pitches in a season. Are there arguments against? Sure, but that's what you all get to fill in.