Burnt Orange Report

The strange thing is, if you look at the polling internals, Kerry slightly outpaces Edwards among Likely Voters, but Edwards outperforms Kerry among Registered Voters and National Adults.

(It's all within the margin of error, but there is a distinct pattern to these results).

This is a little odd - it seems to imply that the sample of likely voters is more keen on Massachusetts liberals than the general population.

Generally, the pool of likely voters tends to be a tad more conservative than the general population (at least in previous USA/Today polls I've seen in past years).

What gives? Are Democrats simply more fired up this year, or did USA Today screw the pooch?

Or am I just nuts?

Jim D,

Something is really weird with the polls coming out recently. Not that I would be thrilled if we were down 12 points, but consider the fact that a week or so ago, Gallup had Bush down 7, then a few days ago it was back to a tie, and now this. That is an awful lot of volatility for a week and a half or so. And as you pointed out, likely voters tend to skew slightly GOP, not heavily Dem as they do in this poll.

Rasmussen has also been bouncing around a lot. He had Bush picking up 6 points on Kerry yesterday, going from a tie to a five point lead. And Zogby just came out with a poll that had Bush leading by 12 points in Red states but behind by 1 point in Blue states. Average that out and it is about a five point lead. But Zogby leans left personally, while Gallup is known for its non-partisanship. They would never deliberately skew numbers to help/hurt anyone.

What gives?

Seriously. Why are the numbers jumping around so much. Unless one of these is the 20th out of 20 polls (you know, accurate with in +- X.XX%, 19 out of 20 times), someone has screwed up badly.

Or polling is a lot less reliable than it once was. I don't know about you, but I will never be polled, since I have a cell phone but no land line, and it is illegal to call cell phones to poll their users. I intend to vote, so that (very very slightly) skews the accuracy of any poll, since I am excluded from the sample. If there are a lot more people with cell phones, but no landlines, or caller ID, or whatever, then polls could be losing their scientific, statistical validity.

Not to say that Bush isn't necessarily down, but just that, statistically speaking, it should be very very rare that two scientific polls from the same distribution have a seventeen point differential, so it is more likely something weird is going on.

Sherk

Here is something to ponder about the Kerry v. Edwards matchup. Kerry forewent federal funding and is thus not bound by federal spending limits. Edwards did not forego the funds as is so bound.

What does this mean? If Edwards is the nominee, he will be hamstrung from Spring until the Nominating Convention in August when Bush (who also forewent federal fundng) will run the biggest negative campaign ever known and Edwards will be defenseless to spend money to respond to it.

Anyone, if I am incorrect, please let me know your thoughts. I like Edwards, but this issue has me thinking.

I have yet to digest the recent decisions about 527s, but those could offset spending limits. However, I'm pretty sure that national candidates were limited in benefitting from the 527s.

But don't forget the lessons of Phil Gramm and Dean: Money isn't everything.

Focus on their messages, on their abilities, on their records, and even on how they might lead the Democratic party.

I think paying too much attention to the money is counterproductive. If I look at non-issue reasons to vote for one or the other, I'd put more emphasis on the appeal to independents, because of their role in a close race. (Edwards does very well among that group.)