Poll of polls has the Zogby pro-Kerry bias at around 1.5%

http://pop.goringe.net/bias.html

I got the following when I did the same process:

Implied one-day figures (Bush)

The worst deviation of the three or seven day figures published from those given by these guestimates is 0.17.

Conclusion - Rasmussen may be dishonest, but not in the way you are saying :)

The poll of polls has Zogby  about 1.5% pro Kerry and Rasmussen about the same amount pro Bush. This isn't about their sincerity, just a recognition that metholodogies have biases, intentional or not.

see - http://tis.goringe.net/pop/bias.html

FWIW, both Zogby and Rasmussen score very well in the 'scatter' of their polls; that is, their results (adjusted for systematic bias) are scattered about the poll of polls mean in the sort of way you would expect from statistical effects.

see - http://tis.goringe.net/pop/scatter.html

My polling rule of thumb is to just average latest Zogby with latest (weekly) Rasmussen.

No, I don't think it was obvious. Seriously. As a liberal Christian, I'm never sure whether non religious people on the left are just using 'Christian' as shorthand for 'religious loony' or whether they really aren't aware that there is are liberal christians. When you are fighting to rehabilitate the word Christian (you know, trying to get it back to something more like 'follower of Christ' rather than 'premillenial fundamentalist nutjob') it's depressing enough when the enemy claim it for themselves - how much more so when you allies accept their definition.

Have you ever felt annoyed that the word liberal has been stolen by the right and made into an insult? Wanted to reclaim the idea that being a liberal is a good thing? Then you know how we feel about the word Christian.

Thanks for inserting the clarifying words into your post.

"And as a Christian, one sees the world through a black and white lens."

"if you are in, you're heaven bound, and if you are not a member of the group, you['re] hell bound"

Please. There are an awful lot of Christians who (a) see the world in shades of grey (b) don't have such an in/out understanding of heaven and hell and (c) don't vote Republican.

Can't post image links here?

http://www.goringe.net/tis/archives/001264.html

I can't fault the analysis (I got similar results when I ran some much more basic calculations), but whatever the polls say, I find FL (98%) hard to believe. We have a 98% chance of winning in FL if an election were today?

(sorry for duplicate: posted this on the poll thread by mistake)

How much time would Kerry/Edwards/their staff have to spend fundraising to get the money? I'd rather they were out in swing states and not asking for cash from certain democrats.

I also think there is a diminishing return: the 10th ad buy has less impact than the 5th, but you can always run one additional campaign trip to meet and greet and generate local press coverage somewhere new.

Add to that the drain on other democratic fundraising, and opting out looks bad to me.

Not so sure... an ad buy during the patriotic fervour that is the olympics might be effective.

Similar question: Can Kerry prepay things like rent? What about wages? I'm sure lots of Kerry campaign employees would happily sign an employment contract which pays them the next 4 months wages in advance....

"Does the guy who compiles this eyewash know the elementary math; that a 1-point lead in such polls could just as easily mean the other guy is leading by 3 or 4 points?"

I hope he doesn't, since that's bollocks.

A MOE of +-4 does not mean that any result in the range is equally likely. It means that there is a normal distribution (or something close to it), centered on the result (in this case +1), and with a standard deviation of 2. There is a 95% chance that the correct value lies within 4 of the quoted figure, but it is not the case that any value in that range is equally likely.

A lead of 1% as measured by a poll with MOE of 4% means there is a 70% chance() the candidate is in the lead. The chance that the other guy is leading by three points or more is about 2.5%.

Sure, we need to treat early polls with caution, but lets get our stats right when we do...

()The formula is complex, but Excel will do it for you: NORMSDIST(x), where x = 2*lead/MOE

From the Things I've Seen poll of polls:

Where do Nader's 3.9% come from?

• 2.8% Kerry
• 0.9% Bush
• 0.2% Other

In other words, according to the polls, Nader isn't bringing anyone from "don't know" or "won't vote"...

Great to see MyDD back again...