McCain had a good debate performance last week, and it seems that some are wondering whether he can make a comeback. Dan Finkelstein of the London Times recently argued:

It's a long shot, I grant, but it does make me wonder whether it's really over for McCain. He is ill-suited to being a front runner but makes a brilliant insurgent. Circumstances have forced him back into that role.

When you look at the Republican race and see Rudy Giuliani out in front, you have to calculate that there is at least one shock out there before we get to convention time.

Why not a resurgent McCain?

Marc Ambinder recently reported on the contents of an internal memo from the McCain campaign. It clearly indicates that McCain has a strategy to do precisely what Finkelstein suggests.

Can this happen?

I think that it could, yes. Certainly, the sense in the Republican Party about the success of the surge in Iraq is something that McCain could use. Generally, the points in the memo Ambinder published seemed reasonable enough. So, I think it could indeed happen - especially if McCain sticks to issues that he is strong on and that engage the Republican base, e.g. spending instead of immigration.

What I wonder about is whether the media will "let him in." Right now, there are already three Republican candidates that the media has in the race - and, oddly enough, there have always been three candidates in the race. Ditto the Democrats. The media consistently has thought about and reported on these nomination battles as if they are three-way races. This has had the effect of enhancing the prospects of John Edwards, who is usually included in the top tier. This enhancement is perhaps artificial, as Edwards' level of support in our RCP national average is closer to Bill Richardson's than it is to Barack Obama's. It has also had the effect of diminishing the prospects of John McCain, who is now usually excluded from the top tier. This has perhaps been artificial as well, in light of the fact that McCain's numbers are hardly different from Romney's in our national average.

Maybe this is not coincidental. Maybe it gets much harder for analysts to do their thing with four candidates - and so they all assume a three-way race. Envisioning how one candidate affects the other two, and how the other two affect that candidate in return, is much simpler than envisioning the same interactions between four candidates. Four people is a lot to keep track of - too many?

I think this makes intuitive sense. We simplify all sorts of things so that we can analyze them more precisely. These are what assumptions are all about. Assumptions reduce the number of factors we have to analyze, so they take us further from reality. At the same time, though, assumptions make analysis easier, and therefore make it more accurate. At least with reasonable assumptions, the benefits outweigh the cost. We assume away factors whose effect on reality is minimal so that we can think about matters more clearly.

Personally, I am "naturally" inclined to assume a three-person race over a four-person race, regardless of whether or not the assumption is reasonable. Adding that fourth candidate makes it difficult for me to think about the ways in which the candidates interact with each other and with the electorate. I start to lose track of things in my mind.

If my subtle, psychological inclination for a three-person race is not unique, then McCain might face additional problems in re-entering the top-tier. If the media is inclined to a three-person race, then what would have to happen is Thompson, Giuliani, or Romney would have to stumble in much the same way that McCain did.