I think it is a very volatile race in these early primary caucus states.

We have not seen a recent poll for NH. The only ones we seem to see is ARG polls.

...Hillary and Obama is going to be hard to keep up after Edwards wins Iowa in a rout.

Whichever one comes in third will get the "What's wrong with her campaign?" treatment.

Overall, the fluctuation we see between many early state polls is probably just as connected to a highly volatile early state electorate as it is to different methodologies.

There's also the simple fact that all the polls come with substantial statistical uncertainties.  Is there any evidence the fluctuations are any larger than one would expect, given the uncertainties of the individual polls?

...Edwards will win Iowa by 20, and that's going to be an earthquake in this race.

The biggest surprise in those polls is the 11 point lead for Thompson in SC.

, Fred Thompson leads his rivals with 27%, followed by Mitt Romney at 16%, Sen. John McCain at 15%, Rudy Giuliani at 14% and Newt Gingrich at 11%.
http://politicalwire.com/archives/2007/0 6/04/obama_thompson_lead_in_south_caroli na.html

The poll averages will show the race about the same as it has been in IA and SC. Edwards may be leading by that much in IA but it appears to be a lot more than the poll averages. The SC numbers are similar to other polls.

are a mixed bag.  I do see Obama ahead in SC, and Edwards in IA and Clinton in NH.  The thing is this do you have enough cash to compete beyond these states and NV?  That is the question for Edwards.

Bad news for Obama is that if Edwards wins Iowa, Barack probably has to come second there. The bump for Edwards in New Hampshire, maybe even enough to win it, would probably mean Obama couldn't recover from finishing third in the first two major contests and would lose his lead in South Carolina. Of course Nevada is a wildcard. Frankly, I still think the winner in Iowa takes the whole enchilada.

I think if Edwards win's IA by double digits he will get a big bounce. If Clinton finishes third she will have to fight for her life in NH.

Although I have never talked with the folks at Public Policy Polling, from memory they use a very similar methodology - ie RBS using past voter history to define the sample.

Certainly there are seniors who still have rotary phones that get excluded, and younger voters are more comfortable with automation, but younger voters are also less likely to be at home than seniors. In my experience, it is more common to see seniors respond in greater numbers than the population demographics, requiring re-weighting.

Now, live interviewers may see an even higher proportion of seniors, due to the factors above, but don't assume that IVR yields a high rate of young voters.

What struck me about these polls is that Obama leads Edwards by about 19 points in South Carolina, Edwards home state?

If Edwards cannot even win the PRIMARY in his HOME STATE I don't see how he is possibly going to win the Presidency.

As a HRC supporter, I do wonder how she could lose Iowa and South Carolina and yet still get the nomination. Given the winner of Iowa will get a lot of free pub, it is vital she finish in the top 2. However, even with losses in Ia and SC, wins in Ca, Nv, Fl, Ny, Nj, etc is dominate for her. Why no polls from NH and the other states, I wonder? Does anybody doubt her dominance in states that Edwards couldn't theoretically file his taxes in?

Per RealClearPolitics, some outfit called Datamar is showing Edwards with a 2-point lead in their Florida polling - 26-24-19 for Edwards-Clinton-Obama respectively.

Yeah, I know - who the Sam Hill is Datamar?  So I'm taking that one with a truckload of salt.  But nonetheless, it's a poll in another early state that doesn't show Clinton in the lead.  So of the five 'early' states (IA, NH, NV, SC, FL), three of them have conflicting poll leads on the Dem side, and of course nobody seems to be polling Nevada.  So that's three out of four.

FWIW, a contemporaneous Strategic Vision poll showed Edwards at 19% in Florida, which is distinctly above his national average.  Wonder if he's hit the airwaves down there or what.  But FL would be a good state for him to target for a win after Iowa.

AL B has taken over...

I am not a polling expert but I think I deserve to be heard about this anyway.

Most of this has been explained to me by my roomate.  I am simply an observer.  I am an Edwards supporter who, through such an uncomplicated method as going on Pollster, has found huge problems with the averages on here.

There is one major problem with the averages on this blog.  They always include both ARG and R2K polls in Iowa averags even though Iowa polling from both sources has been thoroughly discredited.
I understand that the people who run this blog have way too much to do to find out whether or not polling sources can be trusted.

But, at least as far as ARG is concerned, this matter has been brought up over and over by numerous people, myself included.
In fact, every Iowa poll that has shown Hillary in the lead has been done by ARG, with the exception of one, which was done by R2k.

R2K polls once every 5 months.  Their first poll included Gore and is the only Iowa poll where Obama was in first, tied with Edwards.

The only credible Iowa poll that ever produced anything but an Edwards win was a Zogby poll in which he tied with Hillary.

Besides that there have been nearly 20 polls over the course of the last year that have shown Edwards leading in Iowa.

ARG Iowa polling has been thoroughly debunked.  I've had to do it a million times on here and I'm tired of it.  They include too many non-caucus goers, hence the large spaztic swings for Hillary.  Look at where they are compared to all the polling before and after their polls.  You will see how off they are.  Then look at R2K and how they come out of nowhere to fuck up every 5 months.

The opinion amongst polling experrts is that the best Iowa polling is done by the Des Moines Register.  Incidentally, their polling usually gives wider leads to Edwards than other polling.

Over the coure of a year I would expect Edwards to lose at least a few Iowa polls.  I would have no problem with that.  But I do have a problem when 18 credible Iowa polls show him ahead and shows him tied for first and these polls are averaged with crap to muddy the waters as far as who is really ahead in Iowa goes.

The Iowa electorate may be volatile but Iowa Democrats gravitate towards John Edwards.

I still can't get over how few people have seen this.

You can go on my favorite pro Edwards blog (besides Iowa For Edwards and Blogs For Edwards) and get the lowdown on Iowa polling.

When you see all the Iowa polls in a row you see how stupid it is to include ARG and R2k polls in any average.  It's not just that they are "outliers" it's that ARG's screens are too lose...hell why do I even bother.

You cannot look at the pattern (I say "around" becuase I forget the exact numbers she was up by)
and see what is going on.  It looks like ARG tightened their "screen" when they were under pressure from their first two polls being way off and then they either changed back to where the "screens" were or there is something else wrong with their polling.  Why else would they be so off at the start and then all of a sudden come closer to the other polls and then slide back again.  I know that support in polls is fluid but the other Iowa polling is pretty damn consistent.
Do you really believe that Edwards gained 23% on Clinton in a few months?  That means that all the other polls showing Edwards with around a 7% lead where wrong and Clinton really had around a 12% lead?  You realize that if you are including ARG then that is what you are saying.  So after Edwards miraculously gaines 23% worth of support on Clinton she magically takes around 6% of it back?  Why does the Iowa polling look like this...

Edwards ahead
Edwards ahead
Edwards ahead
ARG shows Clinton ahead by around 11%
Edwards ahead
Edwards ahead
ARG shows Clinton up by around 17%
Edwards ahead
Edwards ahead
Edwards ahead
ARG shows Clinton up by 1%
Edwards ahead
Edwards ahead
ARG shows Edwards up by around 4%
Edwards ahead
Edwards ahead
Edwards ahead
ARG shows Clinton up again
Edwards ahead

Why does ARG always show the movement?  I don't  understand all things polling but I understand enough about politics that I firmly believe that Senator Clinton is a lot more popular with casual Democrats than with actual caucus goers.  That is why her national polling numbers are so high.  ARG is not able to correctly exclude non caucus goers so they give Clinton an artififical lead.  Then they try to correct the problem only to settle on some middle ground.  That is the only explanation that makes sense.

I am so tired of the pro-Clinton group who cite polling from these two shitty sources to claim that there is not a consensus among Iowa polls. There is. My brain is fried with anger at this bullshit.

I may be too tired and frustrated to make this clear but go to...

http://democratsunitedforedwards.blogspo t.com/

and read what roomie has to say about Iowa polling.  I am angry because I sense that others are aware of this but for whatever reason include the ARG polling anyway.

When one source of an irregular result is discredited and another rarely produces any result and could likely suffer from the same fatal flaw as the first source wouldn't it be best to not include either or at least provide two averages.

Even when most Iowa polls show Edwards ahead by 6  or 7 or 8 or whatever, including polls showing Clinton up by 11% or 17% or even tied (all from the same screwed up source) are going to wildly skew your averages.

I like this place.  But too many people on here act like the very pundits who most of us hate.
They pick and chose which nuggets of conventional wisdom to swallow and in the end turn into watered down versions of the very observers they often rail against.

I don't have any super duper double secret insider or blogger credential but I can tell that when the same crap comes from the same place...avoid that place.

Here's a few rules that my un-expert as relies on that any average political junkie should follow.  These are not my rules.  They have been taught to me.  And after evaluating the evidence for myself I agree  with the conclusions made.

National polling seems to be worthless, at least for now.  Look at national polling from years past and you will see what I am talking about.
If national polling could somehow magically only include actual primary/caucus goers I think it would look like...

Obama 27%
Clinton - 25%
Edwards - 18%
Richardson - 6%
Biden - 3%
Dodd - 2%
Kucinich - 2%
Gravel - 1%
Unsure - 15%

Just my educated guess. I think that a lot of "insiders" would say that this is closer to reality than some ABC poll and all of it can change after Iowa or New Hampshire or any other major chance at free media.

Don't listen to people who act like they know everything, but if something is evident admit it

I wish people would consider this.

I can admit that Hillary has a lot of money from Wall Street and Johnny Q Billionaire and support from women who see this as an opportunity, and the best hacksters that money can buy...

I can admit that Barack Obama has a lot of love (for now) from the media, he has motivated a lot of people, and he has found a billion different ways to use the word "hope".  I can also admit that he is good at getting people to swallow the idea that he stands for a "new kind of politics"
while having his financial team call the spouses of lobbyists so he doen't break his pledge. I can also admit that he is probably ahead among those who will actually participate nationally right now.

I think a few people on here can admit that John Edwards has won every credible Iowa poll, except for the Zogby poll that he tied for first in.