November 2nd, 2015

#FZGPS: Ann Selzer & Nate Cohn deconstruct the meaning of national polls to the 2017 Iowa caucuses

Sunday’s FAREED ZAKARIA GPS on CNN/U.S. featured an interview with polling experts, Ann Selzer, Selzer & Company, and Nate Cohn, The New York Times.  They discussed the 2016 U.S. presidential campaign polls and what they mean at this stage of the campaign.

MANDATORY CREDIT for reference and usage: “CNN’s FAREED ZAKARIA GPS”



FAREED ZAKARIA, HOST: Trump is soaring. Wait, no, Carson is now ahead. Hillary has a solid lead. But wait, Bernie is rising in Iowa.

Whenever I see headlines about the race and hear news reports about it, I scratch my head and wonder what to think. What should I pay attention to? What should I ignore? And how in the world would anyone know when there are so many Republican candidates in the field?

So I brought in two actual experts to help me understand. Joining me now, Ann Selzer, the queen of polling in Iowa. She runs the polling firm Selzer & Company, based in Des Moines, Iowa. And Nate Cohn writes for the New York Times politics site, The Upshot.

So that first question, Ann, with a huge field like this, I sometimes think to myself, you know, how would one figure out, for example, there obviously is some kind of an anti-Trump vote or non-Trump vote, but it’s probably divided by — among eight or nine candidates right now. So is there some way to figure out what this is going to look like when it winnows down to four?

ANN SELZER, PRESIDENT, SELZER & COMPANY: Well, we have some signals about how that’s going to look. But I agree with you. When we started polling, looking at our preliminary data going, well, how are we going to really sort of set what the tiers are, because you have — can’t really have a — well, you can have a top tier, that’s fine; but your middle tier is gigantic.  And then you have people who don’t even score.

So we were saying, we have to look at more than just the horse race question in order to really understand people’s various strengths. So with a little index we put together called the Selzer Score, we created a way to see Marco Rubio, for example, had upside potential way early, when we first started polling…

ZAKARIA:  Why? So what were you testing for?

SELZER: You know, his number in the first choice was not very big, but he had a very high number for second choice, and had a very high number of people who say, he’s not my first or second choice but I could see myself ever voting for him. And so he rose above where his first place position was because you only get one vote.

ZAKARIA: Do you think that at the end of the day, the Republicans will end up with a more conventional candidate?

NATE COHN, THE UPSHOT: I think so. And I think it’s important to remember that the polls right now do not necessarily reflect where the results are going to be in a few months. If you look back over the last four competitive contests going back to 2004 Democratic race, and then you look at it by Iowa and New Hampshire and the national polls, those 12 contests, the polls right now only get two of them right. They get New Hampshire for Romney and they get Clinton for New Ham — and they get New Hampshire for Clinton. They miss the national result in all of them. They miss Iowa in all of them. A lot is going to change.

ZAKARIA: So in other words, the people who were leading then — at this point in the race, it was all wrong?

COHN:  They all went on to lose. And oftentimes, the people who did go on to win, like Rick Santorum in 2012, were registering at the bottom of the polls.

SELZER: Right.

COHN: So, you know, what we’re seeing right now is that Republican candidates — I’m sorry, Republican voters haven’t really begun to think very much about this race. They’re just now beginning to tune in. They’re going to change their minds a lot. And what we’re seeing is a reflection of the media, of the debates, but not necessarily the underlying factors that will ultimately drive their decisions and the outcome of the race.

ZAKARIA: Can the debates historically — I mean, you’ve been doing this for a long time — can performance in the — in a debate really boost people’s viability to the point that it actually starts registering?

SELZER: Well, the one I’m remembering very distinctly is before I was actually doing the polling. But it was George McGovern in the 1984 debate, who was the only statesman, you know, presenting himself that way and saying, look, if you don’t think I can win, then you vote whoever it is that you want. But if you want to send a message about what you want out of your Democratic candidate, vote for me.

And he surged up to third place. He had been polling very low. So debates do matter

ZAKARIA: A lot of people look at Iowa and say, wait a minute, this rural, agricultural state, full — you know, with huge numbers of Evangelicals, totally unrepresentative of this big industrial — post-industrial country, why the hell do they get to go first?

SELZER: Well, you can make the argument that Iowa is the worst place to start the contest, except for every place else. You’re going to start it somewhere.

The benefit that Iowa has is that you can conquer the state. You can’t conquer California in a meaningful way, in a personal way. You can’t conquer Texas. Iowa, you can drive up through Sioux County and you can go to Allamakee County, and you can meet the people and learn it.

And the point is, you’ve got to organize people to show up. So it really is a test of a campaign’s ability to get things done in a way that isn’t as true in other states.

ZAKARIA: What do you think? What do — what are people focusing on which is the wrong way to think about this race?

COHN: I think people look at who is at the top of the polls and they’re very impressed with Donald Trump having 25 percent of the polls, and that’s just not that much historically. There are a lot of candidates that have been 25 percent, 30 percent in the polls and have gone on to lose.  There have been a lot candidates with a lot less than that who have gone on to win.

And the reality is that it will not take a lot of votes to win Iowa. It is a low turnout race with a lot of people in the field. You only need a small segment of the electorate to ultimately get behind you. I don’t know that the poll — that the way people are reading the polls right now, focusing on a flashy number at the top, is really going to say very much about how it will end up.

ZAKARIA: That’s true on the Republican side…

COHN: Yes.

ZAKARIA: — but on the Democratic side…

COHN: Well, the Democratic side…

ZAKARIA: — tough to…

COHN: Well, the Democratic side may just be as clear as it looks. Hillary Clinton may well be on her way to a very easy victory nationally and perhaps in Iowa.

ZAKARIA: Would you agree — would you agree with that?

SELZER: We don’t know. I may — I never make predictions, because they always include a wish. So I never let it out. We know that you get hot at the end and that’s the way to win the Iowa Caucuses. And we’ll know when it happens.

ZAKARIA: Ann, Nate, pleasure to have you guys on.

COHN: Great.