Polls and the Iowa Caucuses

As an Iowan who experienced the Democratic caucus last time around, I'd like to offer a perspective that sometimes attracts a little coverage (as in this 2004 piece) but then disappears for a long time.

When you go to the caucus in Iowa, the first stage of the process is like a live, public primary: each candidate has a designated space, and his or her supporters go to that place. But then a viability rule kicks in: any candidate with less than 15 percent support (or more, depending on the situation) is declared non-viable, and his or her supporters go other groups.

Based on today's poll numbers, therefore, a typical precinct will see everybody but the Big Three eliminated right away. Your Richardsons or Bidens might survive in a precinct or two, but every precinct will have a significant chunk of voters, probably somewhere between 15 and 30 percent, who aren't able to support their first-choice candidates. If Iowa remains close, those voters could play a large, even decisive, role in determining the state winner. This two-stage process will reward candidates with broad support and low negative ratings--the ones most likely to be the second choice of those Dodd supporters who need to find a new horse to ride on caucus night.

My guess is that the caucus process will result in Obama surpassing expectations based on his pre-caucus poll numbers, but that's just my guess. My main point is that anyone considering the possibility of a candidate catching a wave in Iowa should consider the poll numbers in the context of the caucus process.