In late 2003 and early 2004 Kos and his ilk were convinced that despite sagging support in most Iowa polls Dean had some kind of mystery field operation that would prove all the polls and pundits wrong. Field operations are geared towards effecting turnout on election day - plain and simple. For something like the Iowa caucuses, motivating your supporters and getting them to go to some stranger's house and meet for an hour in the middle of January is the focus of months of planning and phone calls and voter canvassing. But such an extensive effort is rarely made outside New Hampshire and Iowa and even the best field teams aren't going to be able to take a candidate who is 5-10 points behind and put him on top. At best a "good" field operation - like Kerry's was in Iowa and Dean's was in New Hampshire- can help a candidate do a few percentage points better then predicted which matters in early primaries and in a handful of the closest states in November.
But most of the time when you hear officials from campaigns hyping their field work its usually an attempt to assuage supporters who may be getting nervous about their candidates weak standing in the polls.
Also, I want to take issue with Matt's statement about Hillary:
- If I thought I could have an effect, I would spend a good amount of effort to prevent Hillary Clinton from winning the nomination. She's going to be a bad President with poor advisors, she's a weak candidate who makes bad hawkish arguments, and she isn't particularly progressive. Within a year and a half of taking office, her approval ratings are going to be lower than Bush's, as she's going to disappoint liberal Democrats just like this Congress has, and is going to get criticism from an angry right.