A shocking new poll from top-rated pollster Selzer & Co. shows President Obama losing Iowa.
The
poll (611 LV, MoE ±4%, 2/12 - 2/15)
shows Ron Paul in the lead. Yes, Ron Paul.
I have a very hard time believing this poll, though Selzer is a pollster with a good track record. Let's look at the results:
Paul / Obama: 49/42
Santorum / Obama: 48/44
Romney / Obama: 46/44
The poll points to Obama's negative approval rating of 46/48 as some kind of supporting evidence. That's pretty much in line with the national numbers, though, and nobody is putting Obama that far back in the national match-ups. Certainly, no one else is showing Paul with any kind of lead anywhere, however slender. I also think that the author's description of Iowa as a "swing state" that is "critical to Obama’s re-election" is quite an over-statement.
No president has been re-elected with a national approval rating
under 49 percent, according to Gallup polling dating to 1964. It’s a
watershed mark, and about 8½ months from the election, Gallup national
polling Saturday had Obama at 46 percent — underwater.
The Hawkeye State has been awash in GOP messages, and, perhaps as a consequence, it’s in a Republican mood.
We have over a year's worth of PPP polling, both nationally and in multiple States, showing remarkably contrary results to Selzer's poll. Pew Research just put Obama eight points ahead of Romney nationally. Even Rasmussen is regularly showing Obama with leads of anywhere from 4 to 8 points. Fox news showed Obama with a five-point lead over Romney.
What really stands out here is Ann Selzer's pointed insistence that Gallup's favourabilty numbers for Obama are below 49% - which means that Obama loses, in her mind. Further, she points to the large amount of campaign ads from the Republican caucus as a major factor in her position that Obama will lose. So she isn't even pretending that this is some kind of recent development. The Iowa caucus occurred over a month before she took her survey. Selzer is essentially contending that Iowa has gone from a nearly 10-point lead for Obama to becoming a red State in just four years. And she is finding an enormous rise in voters breaking for Paul, though Paul lost the Republican caucus -- third place.
Selzer's reliance on right track / wrong track results is also questionable. Other pollsters have indicated that these lousy figures are more of a reflection of dissatisfaction with the Republican obstructionism and wingnut posing than an indictment of Obama.
On top of that, Selzer seems to be strongly implying here that Obama will lose nationally. What else can be made from her historical comparison and the importance of Gallup's magical 49% mark?
Rasmussen showed
Florida breaking 47/43 in Obama's favour. Did FL not see an enormous amount of Republican campaign ads in their Republican primary? This survey also was a sample of "likely voters", so Selzer can't hide behind that. If Gallup's favourables are such a determining factor, wouldn't Florida be showing up as a sold red State? It didn't break in Obama's favour by almost ten points in 2008, like Iowa did. It broke 51/49 in Obama's favour.
There's also Ohio. Nobody is showing Republicans with a huge lead in that true "swing State", and Gallup is still below 49% for Obama in job approval. Virginia is a State that went for Obama by 6 points in 2008, and Quinnipiac showed Obama with a 4-point lead over Romney in February. But Gallup is under 49%! How can this be, if Gallup's 49% mark is so absolutely reliable of an indicator?
Nothing about her narrative is supported by any polling other than her own single poll. This idea of hers that Obama's job approval is the key is fatally flawed, because Republicans are viewed even less favourably than the President is. Even the "very favorable" and "mostly favorable" numbers that she published don't show a huge advantage for Republicans. Obama's "very favorable" 19% beats all of the Republicans except Santorum, which is a tie. Obama's "mostly favorable" figure of 27% is not substantially different from the other candidates' 30-32%.
Instead, Selzer finds Republican candidates are simply not hated as much as Obama in Iowa. Her entire narrative seems to be built on disliked Republican candidates getting more votes than an even more-disliked Obama.
And she is selling a narrative. The entire article reads like a Republican campaign ad. Her tone is one that presents this narrative as something completely obvious, rather than as a pollster publishing outlier numbers that many would be inclined to be sceptical of. I was almost surprised to find an absence of polling on Obama's birth certificate; it would have fitted right in.
Ann Selzer is sticking her neck out in a big way here. She is seeing an electorate that is far, far more concerned about "fiscal issues" than any other pollster. Maybe she is using a flaky "likely voter" screen that excludes Democrats. I consider it highly doubtful that Iowa is some kind of unique island of public opinion where Ron Paul is seen nearly a dozen points higher than in similar States. That means that, if Selzer is correct, we should see this dramatic flip in many other States. But we aren't.
Minnesota, for example, is Iowa's next-door neighbour. Odd, then, that two pollsters have found Obama comfortably ahead in three polls. Minnesota just went through a Republican caucus campaign, and SUSA didn't find this very strange dramatic shift to the Right.
The 2012 election could be the one that makes Selzer the laughing-stock of the polling world, much like Rasmussen in 2010. Rasmussen also saw a much more "fiscal issue" election and a much stronger Tea Party influence than other pollsters. He ended up with an 8-point bias in the Republican direction.
Selzer has set up a narrative (importance of Gallup favourables, predominance of fiscal issues, Paul cross-over support, rage over the economy, etc.) that runs directly contrary to all other pollsters. She has also boxed herself in to an incredibly tight corner, because her narrative allows no "things changed" moments. The economy will not dramatically improve by the election. The country will still have staggering debt. Republicans will still be running campaign ads criticising the President. Gas prices will still be high. Obama will probably not be above 49% in Gallup's polling. Yet, these are the things she has staked her unique narrative on. If Iowa doesn't go "red" in November, Selzer is left looking like a fool. She can't even hide behind the margin of error, except in the case of Romney, and even then only in a close race.
“Obama’s unpopular in Iowa but he appears to be in position to win
the state next year anyway,” said Dean Debnam, President of Public
Policy Polling. “Extended exposure to the Republican field of candidates
doesn’t seem to have Democrats and independents in Iowa pining to vote
for the GOP next fall.”
PPP surveyed 749 Iowa voters from October 7th to 10th. The margin of error for the survey is +/-3.6%.
Q1 Do you approve or disapprove of President
Barack Obama’s job performance?
Approve .......................................................... 43%
Disapprove...................................................... 52%
Not sure .......................................................... 5%
Q7 Do you have a favorable or unfavorable opinion
of Mitt Romney?
Favorable........................................................ 34%
Unfavorable .................................................... 52%
Not sure .......................................................... 15%
Q11 If the candidates for President next year were
Democrat Barack Obama and Republican Ron
Paul, who would you vote for?
Barack Obama................................................ 47%
Ron Paul ......................................................... 40%
Undecided....................................................... 13%
Q13 If the candidates for President next year were
Democrat Barack Obama and Republican Mitt
Romney, who would you vote for?
Barack Obama................................................ 46%
Mitt Romney.................................................... 42%
Undecided....................................................... 12%
Q14 Who did you vote for President in 2008?
John McCain................................................... 41%
Barack Obama................................................ 48%
Someone Else/Don't Remember..................... 11%
Q5 Do you have a favorable or unfavorable opinion
of Ron Paul?
Favorable........................................................ 27%
Unfavorable .................................................... 55%
Not sure .......................................................... 19%
- PPP shows a lower job approval rating than Gallup did at the time. Yet Obama prevails by four points over Romney, and beats Paul by seven.
- PPP's sample shows a more Republican lean than in the actual 2008 election results, with McCain polling about 3 points better than he actually did. Any idea that the sample was skewed pro-Democrat should be dismissed by that fact.
- No big support for Paul was noted. In fact, Paul was underwater 2-1. Weird, huh?
- Gosh, Romney is even less popular than the 'despised' Obama. Go figure.
- PPP didn't find this weird phenomena of Republican campaign ads turning Iowa into a "red" State.
- Paul showed about the same "cross-over support" as most of the other Republican candidates.
Okay, a four-point lead for Obama isn't some kind solid edge, but it's pretty different than what Selzer is seeing. Both polls are land-line-only. Is Selzer really going to point to her "likely voter" screen as the defining difference?
I'm strongly inclined to think that Selzer is way off base here. She seems to have "gone wingnut", or perhaps seeks to get more business polling for Republicans. It's hard to see this as something other than an ideologically-motivated push-back on the mass of polls showing Obama doing well, but she is kind of restricted to pushing back in Iowa. She has put her reputation on the line, and embraced a faded 2010-style mindset that really no longer seems relevant. She seems to have tapped into a wacky sample, but failed to have the sense to evaluate it.
I will definitely follow up on this as more polling is done of Iowa, and I expect to be merciless on Selzer in November if these numbers of hers aren't backed up by the election results.
How can she get out of this corner she has painted herself into?
If Ron Paul isn't on the ballot, she can claim that her voters were dis-enchanted or discouraged. But this is a sample of likely voters. Either they are likely to vote, or they aren't. Either way, she would have a hard time trying to sell the idea that a whole lot of Ron Paul voters went over to Obama. Because he's so disliked in these survey results. And because everything else (Gallup, right/wrong track, debt, etc.) will still be there.
If it's a real close race (like a recount) and Santorum is the candidate, she can claim margin of error. This would sound a bit pathetic coming from someone with Selzer's reputation, though. Even a narrow two-point victory for Obama would discredit her.
If Romney is the candidate, and Obama wins by any more than two points, she will look stupid and biased. People will point to this poll for the next eight years whenever her name comes up.
The big problem for Selzer in November will be this wingnut narrative she has presented as obvious and self-evident. As I've said, all of these things that she points to now will very likely still be there in November. Even if Obama loses Iowa, but wins nationally, Selzer will look like a fool because of this narrative that sounds like Gingrich on a bender. As goes the Republican Party, so goes Selzer.
What might be most interesting about the Washington numbers is how Mitt Romney stacks up against Obama compared to the other Republican candidates. Long perceived as the most electable GOP hopeful, Romney fares only the third best in Washington with both Rick Santorum and Ron Paul coming a few points closer to Obama.
