Pollster.com Iowa poll trends (nice map)

[Herman Munster]

[smalll]

1. Thanks for the graphs. Clearly, any one of the big three could win Iowa.

I am completely prepared for either Obama, Edwards or Clinton to win there, and will be completely unsurprised when one of them does. I am afraid, however, that many people here on DU are not so well-prepared for all of these eventualities. People need to come to terms with the fact that Hillary could actually win the caucuses. DU might not survive if people don't brace themselves for it.

[Stop Cornyn]

2. Wow, that looks AWFUL for Richardson and Biden. I thought they were moving up, no?

[JDPriestly]

3. The polls don't necessarily reflect how the caucuses turn out, as I understand it.

First, as I understand it, the outcome of the Iowa caucus does not depend who gets the most overall votes, but rather each voting district (is it by county or precinct? or is there no difference?) gets the same number of votes as every other district. So if you have 800 voters in one district and ten in another, the ten and the 800 get the same number of ultimate votes. I might be wrong, but that is how I understand it. Also, if your candidate does not get, I believe 15% of the votes on the first vote, you have to vote another round for your second choice of candidate. That is why, although Edwards does not poll as high as the others, he may well win and he could even win "big."

Frankly, I think the second choice aspect of the whole procedure is very good for the primaries. That mechanism identifies the unity or compromise candidate. It makes it less likely in a field of several candidates such as the field this time in Iowa, a candidate who is intensely disliked by almost as many as he or she is liked by will prevail over a candidate who is maybe not everyone's first choice but who is liked by enough people to unite a majority of the party behind him or her.

We shall see. Anything can happen in Iowa. Yes. I will be extremely disappointed if Hillary wins. I really think it would be a big mistake for Democrats to run on a Hillary Clinton ticket. A huge mistake.

[maximusveritas]

4. Not exactly

Precincts don't get equal numbers of delegates. They are distributed somewhat proportionally based on the number of Democratic votes in the last General Election. The problem with this is that a rural Republican-leaning county will end up with less caucusgoers relative to that proportion. So those who do show up to caucus in those counties will have their vote count more than someone in a strong Democratic precinct. Hope that made sense. It's a strange system that needs to be reformed.

[Herman Munster]

5. yes

what is probably going on in the polls is Edward's support is under sampled and Obama's support is over sampled.

I posted a thread about this. But basically a lot of Obama's young supporters are concentrated in a few counties with college campuses so his votes are going to be worth less on caucus day than Edward's more rural support. This is probably going to surprise a lot of Obama people, but it's the way the caucus system works.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=132x3912775

[maximusveritas]

6. Do you know if any pollsters adjust for this in their polls?

I don't think they do, but I don't know for sure.

[calteacherguy]

7. That bit of a downturn you see for Obama in the "sensitive trend"

That's students home for the holidays.

[Herman Munster]

8. your students are not worth that much

becaus they are concentrated in just a few counties. As I said, Obama's polling is oversampled to begin with. Show me a poll that adjusts for county differences in delegate selection that favors rural counties. I don't know of one.