At what point are we going to start acknowledging that the national and local polls are telling us that it is Pawlenty and Huntsman that are the clearly marginal candidates, and Paul, Bachmann, Cain, and even the ridiculous Gingrich are the competitive ones?Answer? At basically the same point we start taking seriously those polls from a few weeks ago that show Republicans all want to nominate Donald Trump, or the ones from last week showing that Rudy Giuliani is about to capture the nomination. In other words, we don't. The polls just don't tell us very much right now.
They may, to be sure, be telling us something, but only if we interpret them very, very carefully. For example, there's almost certainly something to Sarah Palin's relatively high negative rating among Republicans. But for the most part, remember that early presidential polling is all about asking people who don't have any opinions yet about a question they'll never have to answer.
What do these poll spikes for fringe candidates mean? Steve Kornacki had a very nice item yesterday talking about why someone such as Cain can get a bit of momentum now, compared to Morry Taylor in 1996 (don't remember him? Then you're not as much of a political junkie as you think you are). I think my brother talks about "product lines" that Fox News and others create and sell, and it makes sense in those terms. Fox has 24 hours to fill every day; they don't want to bore their customers with twenty months of Romney vs. Pawlenty. So if they have a month of Trump, a month of Newt, and a month of Cain...well, that's going to produce short-term polling spikes, but it's all going to be forgotten by January 2012.
To answer Larison a bit more directly: right now, I'd almost completely ignore the polls. I'd pay attention to high-profile endorsements, fundraising success, and any other signs of party support -- success in signing up prominent staffers, for example. I'd also pay a lot of attention to anti-endorsements: any strong statements by important GOP leaders that a candidate or a candidate's issue positions are unacceptable (or just the fact of unacceptable issue positions; that's why Hunstman isn't, in my view, a plausible nominee). After the Ames straw poll later this summer, I'd start gradually paying a bit of attention to Iowa polling, and once we're within a month of the caucuses I'd pay a lot of attention to that, and some attention to New Hampshire polling, while still keeping a solid eye on endorsements and other indications of party support.