Tactical Voting

It seems to me that there's no real point in arguing about the significance of the rather large +/- 7 points margin of error on this Newsweek poll showing Barack Obama in the lead for the Iowa caucuses. For something like this, uncertainty about the likely voter screen are probably going to be a bigger problem than sampling error anyway.

But even more to the point, in a close, multi-candidate race the actual method used by the caucuses to allocate delegates starts to make a big different. This method is, especially on the Democratic side, very complicated and tactical voting can start to make a big difference. This is an issue I haven't seem much coverage of, probably because preparations for it on the ground won't start happening until much closer to election day, but one key factor in Iowa is going to be where Edwards and Obama supporters go in caucus sites where they aren't strong enough to win delegates for their guy. Part of what made Howard Dean's task in Iowa so difficult was that almost everyone who wasn't firmly in his camp was firmly against him. In DC, at least, people tend to have Hillary as their first choice or else as a third or lower choice. If that pattern exists in Iowa as well (and I'm not sure that it does) that can be a big problem for her.

Comments

On occasion you have to take the respective operations on the ground into consideration.

Just look who borders Iowa and you'll understand where the poll numbers are coming from.

Posted by leo | September 30, 2007 1:36 PM

I don't think you understand how things work in Iowa.

No on cares how many delegates go to the convention. So all the comments about tactical voting are only important for Hawkeyes.

The national media is going to report the results from the initial round or even the voter preference when the delegates walk in the door.

How many delegates went to Dean in 1984? Who cares? He came in 3rd in the beauty contest and no one remembers what the delegate count was.

Posted by neil wilsom | September 30, 2007 1:38 PM

obama, edwards, and clinton won't have any trouble meeting threshold viability in any of these precincts. it's only 15% of the total precinct vote, and the smaller the community the less raw numbers you need.

the question for tactical voting, is where do biden, richardson, dodd, kucinich and gravel iowa voters go when their guy is below the %15 mark?

I have no idea. I would think hrc would have the fewest potential pick-ups because she's rarely people's number 2 with richardson, dodd, biden etc. believers. Obama seems to be well-liked by almost everyone but will a biden supporter who presumably picked biden because of his experience shift to obama? I don't know.

If I had to guess I would say obama will do well with richardson and kucinich supporters, namely voters who like personality, or obama's antiwar cred or are independents.

I also think he'll do well with edwards supporters who think that edwards can't win the nomination against hillary if it's just him and her. he'd get totally outspent and edwards has shown very little national appeal, something a candidate needs after the first four early states. Indeed, I'm not sure edwards would win in nh against even if he won iowa. if he can't take his momentum past iowa how's he going to win on Feb 5?

Posted by dpg | September 30, 2007 1:49 PM

Neil Wilsom, the only count that matters is the delegate count. The news hardly ever even reports the actual vote count (which was supposedly very close in 2004), they only trumpet the percentages of delegates won.

Posted by Clark | September 30, 2007 1:52 PM

I think Neil isn't right about what is publicised about Iowa. Any results I've ever seen have been the count of delegates elected to the state convention. Certainly the vote that Dean famously came a distant third in was the state delegate count rate. I don't think the count 'as people walk in the door' is even computed, let alone released.

Having said that, I also haven't seen any polling backing up what Matt says about HRC. My impression is that in polling she's often people's 2nd choice. These people tend not to write blogs or live in DC, but I think the population who are happy with HRC but like someone else a bit more is a lot larger than we'd expect from the blogs/DC.

Posted by Brian Weatherson | September 30, 2007 1:57 PM

Tactical voting-wise, both Obama and Edwards have every reason to keep the other from being first, if they are not able to be first themselves. They are better off being second to Hillary than second (or third) to each other. If Hillary does not think she'll win, you know she'll do everything in her power to keep Obama off the top. But what do Edwards and Obama do if they don't think they have the votes? Push for Hillary to be on top?

Posted by cms | September 30, 2007 1:58 PM

In DC, at least, people tend to have Hillary as their first choice or else as a third or lower choice. If that pattern exists in Iowa as well (and I'm not sure that it does) that can be a big problem for her.

To the extent that it's reliable (which is obviously an open question), the polling data answers this for you. 24% name her as their first choice, and an additional 20% name her as their second choice.

Posted by Aaron S. Veenstra | September 30, 2007 1:59 PM

Re Matthew's comment " Part of what made Howard Dean's task in Iowa so difficult was that almost everyone who wasn't firmly in his camp was firmly against him "
-----------
Ha ha ha. What made Dean's task difficult was that billionaire S Daniel Abraham -- a long time advocate for Israel -- threw $200,000 into a 527 to blow Dean out of the water with a series of TV attack ads showing Bin Laden and suggesting Dean was soft on terrorism.

This after Dean argued during a debate with Joe Lieberman that the US should be evenhanded in the Israel-Palestinian dispute. Prior to that, Danny had given $5000 to Dean's campaign.

NO ONE could find out who was funding the 527 until weeks later (because of FEC reporting requirements.)

See "Political Mugging in America" at
http://www.public-i.org/report.aspx?aid=194&sid=200

Posted by Don Williams | September 30, 2007 1:59 PM

the question for tactical voting, is where do biden, richardson, dodd, kucinich and gravel iowa voters go when their guy is below the %15 mark?

This is right. One of the reasons Edwards got so close in IA in '04 is that he got a lot of Gep. caucusers (?) when Gep folded.

edwards has shown very little national appeal

Not right. I wish people wouldn't mix their wishful thinking in with their analysis; it's sneaky and pointless - so Bushish. I do think this commenter is right that Obama could get a lot of Edwards people if Edwards were to fold, but that's very unlikely (weren't we talking about Gravel, Kucinich, et. al.?). IA is going to be close, and the polls mean even less than usual.

Posted by jonnybutter | September 30, 2007 2:05 PM

Since our SLC evidently can't afford a subscription to Forward and Haaretz, I'll pass alone this item from a Feb 2004 article in Forward's "Campaign Confidential" re how Howard Dean got a ..er..extended bris milah.
------------
"Fat Contribution: The contribution of Slim-Fast mogul S. Daniel Abraham to a group that ran negative ads about Dean raised eyebrows in Washington political circles.

Abraham, a major Jewish philanthropist, gave $100,000 to Americans for Jobs, Healthcare & Progressive Values, the Associated Press reported, making him one of the two biggest donors to the group.

The group ran ads attacking Dean in Iowa, New Hampshire and South Carolina, including one raising questions about his lack of foreign policy experience that showed the face of Osama bin Laden. Its funders included supporters of Gephardt and Kerry.

What raised eyebrows, however, is that Abraham’s longtime chief political factotum, Sara Ehrman, was an adviser to Dean.

Ehrman, who works as senior adviser at the Abraham-funded Center for Middle East Peace and Economic Cooperation, declined to comment on Abraham’s funding of the anti-Dean group or how her involvement with Dean might be viewed in light of it, saying only, “Danny’s going to have to speak on that himself.”

Ehrman, a relatively early supporter of Dean, helped him navigate the shoals of Jewish and Middle East affairs. Given her reputation as a noted dove, Ehrman’s presence on the campaign helped raise suspicions about Dean among hawkish pro-Israel Democrats.

Dean’s policy adviser, Jeremy Ben-Ami, a friend of Ehrman’s, said he had “no clue” why Abraham supported the anti-Dean group, adding, “I don’t know what he was told on being solicited.”

Abraham was traveling and not available for comment."

Ref: http://www.forward.com/articles/campaign-confidential-85/

Posted by Don Williams | September 30, 2007 2:34 PM

They are better off being second to Hillary than second (or third) to each other.

I disagree with you here. If Hillary comes in 3rd then it's a big blow to her inevitability strategy. Obama and Edwards could both claim to have bested the media favorite.

If Obama comes in 2nd to Edwards, he can argue Edwards went all in for Iowa. Barack still has the money and organization to move forwards in later states.

If Edwards comes in 2nd to Obama, he's the comeback kid and can lobby that into support in the later states.

Posted by Jinchi | September 30, 2007 2:36 PM

While Howard Dean was pulling S Daniel Abraham's knife out of his back, he got another one stuck in by his OWN CAMPAIGN MANAGER --Steve Grossman, former chairman of AIPAC. ha ha ha

Right before the pivotal Wisconsin, Grossman announced (falsely) that Dean was throwing in the towel -- to aghast protests from Howard Dean!!! Ha ha ha. Nothing like having a subordinate surrender on your behalf --and without your knowledge -- right before a big battle.

See http://www.thecarpetbaggerreport.com/archives/1259.html

An excerpt:
" Yesterday, in the middle of the day, the AP started reporting that Dean was “prepared to abandon the race,” based on Grossman’s conclusions.

“I have no doubt he’ll support the nominee in any way he can, no matter who the nominee is and obviously that nominee looks to be John Kerry,” Grossman said of Dean. “He may say that Tuesday night. He may wait until Wednesday or Thursday to say that.”

Quickly thereafter, the campaign started challenging Grossman’s comments, saying that Dean would pursue the nomination with or without a victory in Wisconsin.

“We’re not dropping out after Tuesday, period,” Dean said in a television interview yesterday.

Grossman, almost certainly aware of the consequences, then told the New York Times that he would leave the Dean campaign and move into Kerry’s camp if Dean lost in Wisconsin.

“If Howard Dean does not win the Wisconsin primary, I will reach out to John Kerry unless he reaches out to me first,” Grossman said. “I will make it clear that I will do anything and everything I can to help him become the next president, and I will do anything and everything I can to build bridges with the Dean organization.”

None too pleased, the Dean campaign decided to make Grossman’s transition a little easier by announcing that he has, in effect, been fired. How exactly? The campaign issued a press release thanking Grossman for his service. It was written entirely in past tense."

Posted by Don Williams | September 30, 2007 2:49 PM

I wish I knew more about the demographics of the Iowa caucusgoing population. It's usually less than 10% of registered voters. While my sense is that Weatherson is right, above, about a lot of people in the general population picking Clinton second behind Edwards or Obama or whoever, caucusgoers are probably high-information enough that the Clinton name recognition advantages causing this are dampened.

Posted by Neil the Ethical Werewolf | September 30, 2007 3:44 PM

Okay, it seems like nobody here has ever actually organized on the ground for an Iowa caucus. The big count is after the reconfiguration. This is going to be complicated and I'm trying to remember from a quarter of a century ago, but it isn't the actual delegates you send either. What is important are the percentages from the caucuses - in other words the percentage of people who votes for you in each particular caucus as related to the amount of delegates that caucus can give you.

Who is the second choice is really not that big an issue. The much more important issue is who you have on the ground after the first vote and the non-viable groups have to pick new groups. Here is the way it was explained to me - you bargain anything you can to get the non-viable groups to join you. Here is the way the bargaining usually works - let's say there are 100 caucus goes 40 go for candidate A and 50 go for candidate B. Candidate C has ten. Candidate C knows he doesn't have any chance of getting delegates as it is, but the Candidate C leader wants to go to the convention. The leader of the other two groups bargain with candidate C (there is a short time to do this in). You promise your delegate spot (or one of them) if C joins your group. Now you have to have a lot of trust for people in C to believe you will come through, and the people in A to agree to give up a place at the convention.

This is what I learned the hard way in 1984 and Dean learned the hard way in 2004.

Posted by Wilbur | September 30, 2007 4:01 PM

What is the point of having a poll with a margin of error of +/- 7%? All the polls beforehand show a statistical tie, so what new info does this add? What a waste of money. If you're going to do a poll, do it right.

Posted by Reality Man | September 30, 2007 4:49 PM

Welp, since the "Big Three" candidates have pretty much said that they won't be ending our glorious Iraq adventure, seems to me that the only sensible thing to do is to tell Beltway ladder climbers to shove it, and vote on principles. Which pretty much means Kucinich, now, for me. In the general election, a vote pretty much means fuck-all. I don't know if I'll bother in '08.

After the Jackass wins its fabulous cathartic victory (assuming no Iran war -- which the Jackass will do nothing to stop -- intervenes), you'll see your health care "reform" written by and for the insurance industry, and your "peace" party will sign off on its own half-trillion dollar "defense" budget. Everything will be different!

Posted by sglover | October 1, 2007 1:52 AM

"I disagree with you here. If Hillary comes in 3rd then it's a big blow to her inevitability strategy. Obama and Edwards could both claim to have bested the media favorite."

Whoever comes in third is very likely going to find themselves in a sudden death spiral. And that includes Hillary.

----

"caucusgoers are probably high-information enough that the Clinton name recognition advantages causing this are dampened."

Yup. This is why Iowa is a three way race instead of a two way race.

Posted by Petey | October 1, 2007 2:12 AM

I'm really glad you brought this up. I've been saying Iowa should never be the state that decides who the nominee is, which is what results with the tight compression of primaries that immediately follow.

It's a downright fruity system, and you're right that it totally killed Dean. His raw volunteers didn't know how to campaign up there.

This time around I'm hoping it plays into Obama's favor (in which case you'll hear me praising that fruity system!).

Another consideration is, how much will the fact that there will undoubtedly be a Republican upset on the same night take away any oxygen from the Democratic winner's momentum?

Posted by KathyF | October 1, 2007 7:28 AM

I am an Iowan and a caucus veteran.

The key factor isn't who Obama and Edwards supporters will go to in the precincts if they don't make the first round cutoff of 15%. The big question is where Richardson, Biden, Dodd, et. al. supporters will go when they don't make the cutoff? There will be very few precincts where Clinton, Obama and Edwards do not make the 15% cutoff. But in almost every precinct there will be 15-25%% of the attendees that night whose candidates will not be viable after the first round. Who will their second choice be?

That is the critical question upon which the night's results will hinge.

Posted by Connor Anderson | October 1, 2007 10:06 AM

If Clinton comes in third in Iowa - no matter how close the numbers - it's HUGE for Obama. What I believe he needs to prove is that his ability to get people to come out and vote for him, not merely send him money. This is particularly crucial amongst African-Americans who would love to vote for him but feel that Clinton has the numbers on her side. An Iowa win would probably guarantee him South Carolina.

If he does well - at least second - in New Hampshire (especially if he brings in independents), first or second with Edwards in Iowa, and wins South Carolina, momentum will seriously be on his side. Clinton won't be out - with all that cash, she'll hold out for wins in the big states - but she'll be seriously wounded. It will be Obama's make-or-break moment to prove he's truly the candidate of change.

His one major mistake recently - not casting a vote in Kyl-Lieberman. By failing to do so, he's ceded a major campaign weapon.

Posted by GordonsGirl | October 1, 2007 11:38 AM