I'd like to believe it, but there was also a U of Iowa poll that showed Obama WAY back in third. I've rarely seen the caucuses polled accurately.

Whoever finishes a distant third is going to be in big trouble.

Obama is doing strong, but with the wrong folks.  He has greater strength among independents and Dems at large, but crashes among probably caucus-goers.

His organization will need to be top-notch to pull off a win.

They probably measure whatever there is to measure fairly accurately.  But it looks like all three frontrunners are within the MOE of each other, and nobody else is in the race.  This is pretty the same result you would get if each of the Iowans polled simply flipped a three-sided coin to pick a candidate.

I think what most of the polls are measuring right now is a whole bunch of nothing.  However soon the candidates announce, however soon they start campaigning, the voters won't engage until they are ready to.  And I don't think they are ready.

Sorry, but Iowa caucus polling is not all over the place.  I understand how it could look that way if you include the bogus American Research Group polls though.

Edwards has won every credible Iowa poll, all 10 of them (he has tied for first twice, once with Obama in the only Iowa poll to include Gore and he tied once with Clinton after her announcemnet).

Before you mention the American Research Polls that show Clinton winning you should know that they have been discredited.  Edwards wins first 4 or 5 Iowa polls.  Then ARG releases their first poll, it gives Clinton a 21 point bounce compared to other polls, placing her in first.  The media declares her the new frontrunner in Iowa.

Then Edwards wins the next two non ARG polls.  Then Clinton wins the second ARG poll, again recieving a 20 point bounce.  James Carville and Team Hillary show up everywhere, taking Hillary's position in the last non-ARG poll before the second ARG poll (down by 10 points) and adding her position in the second ARG poll up by 20% to claim that she got a 30 point bounce from one appearance in Iowa.

Both Obama and Clinton's camps have tried to spin Edwareds consistently winning in Iowa by claiming that he is only ahead because he has been there far more often (which he has) but that isn't the sole reason. Iowa Democrats gravitate to Edwards.  That is proven when the non ARG polls following the second ARG poll still have Edwards in the lead.

At this point people start to take notice.  Pollster.com has some interesting stuff about this.  ARG, though generally credible, did not have tight enough screens.  Too many non-caucus goers were included in their polls.  You know, more casual Dems (nothing wrong with that, we were all there at one point) who answer "Hillary" or "Obama" because those are the two names they hear the most, but it is very unlikely that they will caucus when the time comes. I guess ARG tightened the screens on their third Iowa poll (they are still being ctiticized for not going far enough though) and it showed as Clinton only lead Edwards by 1% point.  

Think of how this relates to the national polling.  "Registerd voters" who either say they are Dems or lean towards Dem candidates are not going to reflect primary/caucus going Democrats.  Something like 6% of self-identified "liberal Democrats" approved of Bush's job performance recently in a major poll.  Bush's approval ratings at a caucus in Iowa would be 1% if he was exremely lucky.  

Take Gore out of the national polling because he's not going to run, figure that his support will likely split 60 - 40 Edwards/Obama (when Edwards goes up, Gore usually goes down, or vice verse, it's happened too often to be a coincidence), and figure in how soft support for the media candidates is, and the difference between their sample and real primary voting Dems, and it's not a stretch to say that if there was somehow  a magical poll that could identify primary Dems (when asked many people lie because they are intrigued) Obama would probably be in the lead but his suppoort would be soft, but not as soft as Clinton's.

The point is that when pollsters like ARG make mistakes, especially with a small sample, it can drastically alter the results.  

I can see how, looking at the ARG polling mixed in with the other Iowa polling it would look "all over the place."  If it had not looked that way people would not have cheched it out to see if something was wrong.  

Here are all the Iowa polls (without the ARG ones of course).  I did not include last summer's Iowa State Fair straw poll in which Edwards and Clinton tied at 30% apiece.

It's also worth noting that while Vilsack did endorse Clinton, most of his machine has gone to Edwards.  The poll that I include at the top shows that the announcement about Elizabeth's cancer coming back (please pray for/ think about her, depending on your beliefs - people always compliment those facing cancer but Elizabeth truly is an amazing, inspirational, woman)hurt Edwards with men but once they see that he is not dropping out it will be unlikely tht they will  let his wife's illness be the deciding factor.

If you think that these polls don't show a definite trend in Iowa then I don't know what to say.

IOWA CAUCUS POLLING

Zogby Poll - March 26, 2007
4.5% margin of error

Likely Caucus Goers

John Edwards - 27%
Hillary Clinton - 25%
Barack Obama - 23%
Joe Biden - 3%
Bill Richardson - 3%
Dennis Kucinich - 1%
Chris Dodd - 1%
Not Sure -15%

University of Iowa Series of Polls  
March 19-25, 2007

Note: These polls measured the differences in support for the candidates before and after Senator Edwards announcement that Elizabeth Edwards had a reacurrence of cancer. They also measured the difference in support between all Democrats and likely caucus goers. This is important to pay attention to because it is another example that Senator Clinton is more popular among more casual Democrats while Senator Edwards is more popular among those who will actually causcus or vote in their primary. It is also worth noting that the only three candidates included were Edwards, Clinton, and Obama, so the results could have been skewed. Among both Clinton performs better than in other recent Iowa polling, while Edwards, and especially Obama, perform worse. The poll is still useful in illustrating that Senator Clinton is not anywhear near as popular with the Democratic base, as most people asssume.

Caucus Goers Only
Before the Edwards's announcement

John Edwards - 30.2%
Hillary Clinton - 24.4%
Barack Obama - 22.1%

Caucus Goers Only - After the annoncement

John Edwards - 36.4%
Hillary Clinton - 33.9%
Barack Obama - 14.4%

All Democrats - Before the announcement

Hillary Clinton - 21.1%
John Edwards - 19.7%
Barack Obama - 18.8%
Don't Know - 32.9%

All Democrats - After the announcement

Hillary Clinton - 29.5%
John Edwards - 23.2%
Barack Obama - 11.4%
Dont't Know - 28.7%

SOMETIME IN BETWEEN THESE POLLS (I forget exactly when) THE THIRD ARG POLL IS RELEASED IT SHOWS CLITNON WITH A 1% LEAD OVER EDWARDS

Strategic Vision Poll - Feb. 16-18, 2007
4% margin of error
Likely Iowa Caucus Goers

John Edwards - 24%
Hillary Clinton - 18%
Barack Obama - 18%
Tom Vilsack - 14%
Joe Biden - 5%
Bill Richardson - 3%
Wesley Clark - 2%
Chris Dodd - 1%
Dennis Kucinich -1%
Undecided - 14%

Zogby Poll - Feb 7-8, 2007.
4% margin of error
Likely Iowa Caucus Goers

John Edwards - 24%
Hillary Clinton - 24%
Barack Obama - 18%
Tom Vilsack - 9%
Joe Biden - 4%

SOMETIME IN BETWEEN THESE TWO POLLS (once again, I forget exactly when) ARG RELEASES THIER 2ND IOWA POLL, IT SHOWS CLINTON AHEAD BY 20 POINTS!
(36% to 16% I believe.)  ARG's METHODS ARE QUESTIONED.

Strategic Vision Poll - Jan. 19-21, 2007
4% margin of error
Likely Iowa Caucus Goers

John Edwards - 25%
Barack Obama - 17%
Tom Vilsack - 16%
Hillary Clinton - 15%
Joe Biden - 4%
John Kerry - 3%
Wesley Clark - 2%
Bill Richardson - 1%
Chris Dodd - 1%
Dennis Kucinich - 1%
Undecided - 15%

Zogby Poll - Jan. 15-16, 2007.
4% margin of error
Likely Iowa Caucus Goers

John Edwards - 27%
Barack Obama - 17%
Tom Vilsack - 16%
Hillary Clinton - 16%
Joe Biden - 3%
John Kerry - 3%
Dennis Kucinich - 1%
Bill Richardson - 1%
Not Sure - 13%

SOMETIME IN BETWEEN THESE TWO POLLS (of course I forgot exactly when) ARG RELEASES IT'S FIRST IOWA POLL. I KNOW THAT HILLARY WINS, I THINK SHE RECIEVS 30% AND EDWARDS GETS 21%.

KCCI-TV Poll - December 22, 2006.
5% margin of error

John Edwards - 22%
Barack Obama - 22%
Tom Vilsack - 12%
Hillary Clinton - 10%
Al Gore - 7%
John Kerry - 5%
Wesley Clark - 4%
Dennis Kucinich - 4%
Joe Biden - 1%
Evan Bayh - 1%
Bill Richardson - 1%
Undecided - 11%

Note: Most Iowa Polls do not include Al Gore, as he is not likely to run. His inclusion almost certainly skewed the result.

Des Moines Register Poll - October 12 - 19, 2006.
4% margin of error
Likely Iowa Caucus Goers

John Edwards - 36%
Hillary Clinton - 16%
Barack Obama - 13%
Tom Vilsack - 9%
John Kerry - 6%
Joe Biden - 5%
Undecided - 8%

Des Moines Register Poll - October 12 - 19, 2006
4% margin of error

Iowa Democratic Party County Chairs and Vice Chairs only

Edwards - 40%
Vilsack -15 %
Obama -11%
Clinton - 8 %

Des Moines Register Poll - June 2006_
4% margin of error
Likely Iowa Caucu Goers

Edwards - 30%
Clinton - 26%
Kerry - 12%
Vilsack - 10%

You do have a point about the influence that the eldelry have in the Iowa caucus but I am a little confused by your statement that...

"...the Iowa poll showed Obama within a reasonable distance among ALL Dems in Iowa, but way back among 'likely caucus goers'. I think that's because caucus goers tend to be, among other things,  much older than most other voters."

This is confusing because "caucus goers" are the only "voters" that matter in Iowa or any state for that matter, but especially in a caucus situation.  Before I get into why it's worth pointing out that reading too much into 1 poll isn't good.  Looking at the totality of Iowa polls paints a much clearer picture.  It's just like the 2008 head to head general election polling.  Hillary does best against the GOP in some polls.  Obama has his moments.  But if you look at the totality of head to head general election polling it becomes clear that Edwards is the most electable Dem.  If you want to see the polling (I won't paste it hear becomes it is far too long) then e-mail me at...

progressive_populist_4edwards@yahoo.com

Back to the caucus.

The winner of the caucus will have the best organization and the most committed supporters.  First of all, unless something drastic happens, it does not seem likely that anyone other than "the big 3" will recieve the 15% necessary to reach "viability".  Especially with Vilsack gone.  With him gone it will now be a 3 person race, and his endorsement is not worth much, many of his supporters were with him for loyalty sake, that loyalty. doesn't transfer to Hillary.  

The voting to reach "viability" can be deceptive.  Edwards, Clinton, and Obama will reach 15% right away.  That's when the Richardson people will try to get those who were going to support one of the big 3 to back him instead, "for the sake of Democracy". Maybe Richhardson has a chance at reaching 15% but I doubt it.  The Biden, Dodd, Kucinich, and Gravel supporters (does Gravel have suppoters? gotta love his gadfly appeal though)will likely be looking for a winner.  They aren't going to jump on the Richardson express.  In 04'Kucinich told his supporters to back Edwards if Kucinich didn't make it to 15%, similar deals could happen this time.

This is why I think the order will be Edwards, Obama, then Clinton.  There is a stronger (and growing) ABH sentiment in Iowa.  Richardson, Dodd, Biden, and Kucincih backers will split pretty evenly towards Edwards and Obama (Obama getting the most) with Clinton getting only a few. By then I think it will be clear that she is unelectable and about as pretentious and calculating as it gets.  

Edwards will have enough rabid supporters.  David Yepsen, the guru of Iowa politics who rights for the Des Moines Register mentioned that it felt like Iowa was holding it's breath in the half a day in between the announcement that Edwards would make an announcemnet, and the announcement itself.  Almost losing his candidacy has lit a fire under Edwards supporters (notice the recent uptick in endorsements, especially in South Carolina).  During the caucus there will be some passionate arguments and it seems that Clinton's supporters are either brain washed and oblivious or on the fence.

Hillary's camp was glad when Obama got in because they wanted the progressive / anti-war / economic populist vote split 2 ways.  You can bet that they love the fact that Al Gore is included in national polls. But if, god forbid, Edwards is forced to drop out, then it's all Obama.  As long as Gore were to stay out Obama would have it in the bag.  As about as serious of an Edwards supporter as it gets I can say with certainty that if he were to drop out (which I doubt he will) then I would back Obama immediately. If the big 3 is still the big 3 in mid January, and if nothing major happens then when it's said and done Iowa will look something like this (for the sake of argument I'll pretend that no one will abstain)...

1st: John Edwards - 40%
2nd: Barack Obama - 30%
3rd: Hillary Clinton - 29%