I've mentioned a few times already that I don't trust Zogby's numbers for the last two days. More to the point, I trust his numbers, but I don't think they're very reliable and I don't trust the conclusions that he draws from them. Read on to see why this is, and what conclusions I draw from his numbers.
What James Zogby is trying to do with his latest poll numbers is capture the movement of the undecideds and "leaners." Doing this requires some creative poll construction and some fudging of the numbers (I'm not sure what the technical term is, but computational math geeks like me call it extrapolation). His numbers from last night show a huge swing in the direction of John Kerry, mostly on the basis of these leaners changing their minds about what's important in the campaign. Zogby concludes that Kerry is going to win in a blowout because ultimately electibility is what matters.
The problem is that Zogby is relying on a snapshot of a very volitle field. Just as those leaners changed their minds suddenly in the course of 24 hours, they could change their minds again overnight. The fact that there was a major shift in Kerry's favor last night after a major shift in Dean's favor the night before, is an indication of just how volitle the field really is. What those numbers really indicate is that there's a huge segment of the New Hampshire electorate -- larger than anyone thinks -- that still hasn't quite made up their minds. Chances are they're not going to know who to pull the lever for until the last second.
So, what conclusions can we draw from this?
- The New Hampshire electorate is conflicted on what they value in a candidate. They admire Dean's vision and values, and feel some sympathy toward him because they resent being manipulated by the media. They value Kerry's experience and heroism, and see him as a safe candidate. Ultimately this is a battle in their hearts and minds between a bold vision and comfort food. New Hampshire still hasn't made up its mind about which is better. But they have made up their minds about which candidate represents which value.
- The campaign is still on in New Hampshire, and the media still plays a huge part. A favorable or unfavorable revelation right now can still swing the outcome, even though polls have been open for six hours.
- Any prediction of the outcome at this point is practically worthless, the same way that a weather report that is accurate to within 30 degrees Farenheit is worthless. I have no reason to doubt the margins of error for the polls we have for the times that they were taken. However, the electorate is simply too unstable to accurately predict their course.
In short, right now, anything is possible.