OK, I’ll get to the links in a few minutes, but I’m riled up about all this fifth-starter stuff that seems to be the talk of the Internet today. Howard Megdal wrote a column for SNY.tv advocating Nelson Figueroa for the fifth spot in the Mets’ rotation.

Sam Page responded to my post from yesterday and Howard’s column, among others, with a comprehensive post at Amazin’ Avenue laying out a series of scenarios. He ultimately concluded that the most competitive rotation for the Mets would involve sending Oliver Perez to the bullpen something he and everyone else knows they won’t ever actually do.

Meanwhile, at Fangraphs, Marc Hulet suggested a new approach to the fifth-starter’s role. He thinks teams should employ a “three-man job share,” where three different pitchers rotate in the fifth spot in the rotation.

In turn, Rob Neyer weighed in on Hulet’s plan and pointed out that it’s a bit specific to the Blue Jays — the example he uses — and that, though the current system of trotting out unqualified retreads is certainly flawed, it’s not clear Hulet’s is markedly better.

Last night, before I read Hulet’s post or Neyer’s response, I began a post about a completely hypothetical idea for the Mets based on something I noted in this item on The Book Blog.

I don’t have The Book handy, but the e-mailing reader points out that “MGL states that most teams would be better served to use three long relievers instead of their fifth best starter.”

It makes some sense, based on the premise that batters get progressively better with increased appearances against a particular pitcher in a game.

It’s not going to happen for the Mets out of the gate, and I’m not certain it should. The team will certainly give Johan Santana, John Maine, Mike Pelfrey and Perez spots in the rotation, and the guy who appears most likely to win the fifth spot, Jon Niese, also appears reasonably likely to be better than all but Santana.

But once one of those guys suffers an injury or proves ineffective, the Mets could feasibly have Figueroa, Hisanori Takashi and Fernando Nieve on the squad as potential replacements, assuming they follow my suggestion, which they won’t, which is part of what makes this completely hypothetical.

None of the potential starters would be ready to throw deep into a game, so, in reality, the Mets would likely ask one to start and pitch a few innings and work on stretching him out in subsequent starts.

But what if they didn’t?

What if, for one turn in the rotation, the Mets just used three relievers?

In theory, it would maximize the effectiveness of all three in that game, since likely none would have to face the same batter more than twice. And it would keep all three fresh enough to allow them to maintain their bullpen roles during the rest of the rotation, so the Mets could essentially have eight relievers while only carrying 12 pitchers. Plus it would guarantee regular work for the long man in the ‘pen, probably an added benefit.

It would require deft management of the bullpen, and I’m not sure Jerry Manuel would be the guy to handle that. Plus, the reliever rotation spot would almost certainly have to be sandwiched by the Mets’ two most reliable starters — Santana and someone — since the bullpen would be shorthanded the day before and after.

It’s not going to happen, but I wonder if it would be more effective than simply inserting one of Figueroa, Takahashi or Nieve into the rotation. I imagine the plan’s downfall would actually be the rest of the Mets’ rotation, which probably — and depressingly — isn’t good enough to endure a couple of days a week with a short-staffed bullpen.