Rove’s Map, Flawed Poll Samples And A Potential Blowout In November

There are tea leaves to be read out there, and there is a message those tea leaves are delivering if you know what to look for.

The message is that Barack Obama is going to be clobbered in November.

First, let’s look at a few indicators of the strength of Obama’s campaign, shall we?

Just from this week’s headlines, we see…

In other words, the 5-to-1 funding advantage Obama had over John McCain four years ago is gone. He will, in all likelihood, be at a disadvantage money-wise.

And what that means is that, barring events helping him or major mistakes by Mitt Romney’s campaign – something Romney’s people have not made many of since the primaries got rolling – this is as good as it’s going to get for the president.

How good is it? Well, InTrade still expects Obama to win – but he’s at 53.4 percent today. That’s the worst it’s been since last fall, before the GOP primaries got going and the candidates beat each other to a pulp. Obama rose into the low 60’s while that was going on but he’s been dropping ever since.

And the RealClearPolitics poll average gives Obama a sliver of a lead – 46.8 percent to 44.2 percent. But the polls they’re using to make that average should be examined, because there’s a consistent issue with the samples in them which invalidate much of their conclusions.

For example, the Rasmussen poll conducted over the weekend has Romney up by a 47-45 margin. It doesn’t have a poll sample listed, but Rasmussen’s samples usually track the latest numbers he has on party identification. The 2010 electoral results showed an electorate which was 35 percent Republican, 35 percent Democrat and 30 percent independent. There isn’t really a reason why you’d depart a whole lot from that number; Rasmussen’s most recent party ID figure is 36 R, 34 D, 30 I.

Gallup’s latest poll has Romney with a 46-45 advantage. That poll doesn’t quite give out its sample, but the weekly demographic master file downloadable from Gallup’s site shows a 35R-32D-33I sample which is very close to the one Rasmussen is using.

The other polls in the RCP average? Well, there’s a Bloomberg poll that has Obama up 53-40, which is an outlier and a laughable one at that. Bloomberg’s poll sample was 33R-38D-26I, which is a skewed sample they’re actually bragging about having been a five-point Democrat edge which is less than the other poll samples of late.

And Bloomberg is right, though it’s clear nobody believes the president is up by 13 points. The Associated Press/GfK poll from last week which had Obama with a 47-44 lead on Romney carried a poll sample that is 22R-29D-33I, which is a D+7 sample that looks just like the 2008 election. If you believe the electorate will look in 2012 like it looked in 2008, maybe you can believe that poll.

Pew Research had a poll with Obama leading Romney 50-46 that was concluded on June 17. The poll sample in that survey was 38R-40D-21I, though rather than ask whether the respondents were R’s, D’s or I’s they asked whether the respondents were “certain” R’s or D’s, or “swing voters.”

Reuters-Ipsos released a poll on June 12 which had Obama ahead of Romney 45-44, with a poll sample of 38R-47D-15I. That’s a D+9 poll.

Monmouth University’s June 7 poll had Obama with a 47-46 advantage, with a poll sample that was 30R-35D-35I. That’s a D+5 poll.

The June 5 Fox News/Opinion Dynamics poll which had the race at a 43-43 tie didn’t show a poll sample.

And the June 11 Investor’s Business Daily/Christian Science Monitor /TIPP Poll which had Obama ahead 46-42 had a poll sample which was 29R-34D-32I. That’s a D+5 poll.

In other words, Rasmussen and Gallup, who tend to be the most accurate pollsters in the industry, both have the true party ID at R+2, but the rest of the polls have samples which are anywhere from D+5 to D+9.

That’s about a seven-point swing in the samples. Let’s discount the swing down to, say, four points and say that on average, Obama is overpolling by four points. Admittedly, Rasmussen and Gallup have the race as Romney +1 or Romney +2 with what we’d call an “accurate” poll sample, but we’re trying to construct an average here – and a four-point swing based on an average oversampling of Obama’s vote seems reasonable.

Now, let’s look at the electoral map Karl Rove has, based on an average of polling the race on a state-by-state basis…

You’ll see in that map that Obama has 194 electoral votes in states where he’s ahead by more than eight points. That’s 13 states, plus the District of Columbia. Some of those states, specifically New Mexico and New Jersey, might be a bit more in play than it currently looks – but we’ll go with those numbers right now.

But Obama also has 63 electoral votes tied up in six states where his lead is between four and eight points. Apply our four point oversample correction, and 39 of those electoral votes – those in Wisconsin, Oregon, Nevada and Michigan – become toss-ups. Obama keeps 24 of those electoral votes, but all of a sudden he looks like he’s really got about 218 votes on his side with 39 in the middle.

Push that four percent to Romney and some states which are currently rated as toss-ups all of a sudden are not. Romney adds 25 votes from Missouri and North Carolina, states he’s already ahead by three points in, and likely another 35 votes from Florida and Iowa where the race is shown as tied. That would put Romney at 232 votes – 14 ahead of Obama.

And the states in the middle? The aforementioned Wisconsin, Oregon, Nevada and Michigan, plus Colorado, Virginia, South Carolina and Ohio. That’s 78 electoral votes in the pot by these calculations.

So if you move numbers by four poll points to account for the oversampling of Democrats you’ll see that Obama is actually DOWN 14 electoral votes with a lot of unfriendly territory to pick up.

If the election was today, Romney likely wins.

But it’s not today.

Things can change between now and Election Day. The money doesn’t look like it’s going to favor the president from here on in; he’s even starting to struggle in Silicon Valley where the lefty-crunchy entrepreneurial set has held sway for decades.

Events might, one supposes. But what events would give Obama a boost he doesn’t currently have?

The economy? That doesn’t seem likely – consumer confidence stinks, and there isn’t any reason to expect it to improve. Europe’s crisis isn’t likely to get better by November, and that’s going to be a drag on our economy. Real estate is still a mess, the EPA and the Interior Department are still doing a number on what domestic energy could do for us, Moody’s is downgrading the banks and in so doing proving that Dodd-Frank was a bust in improving the health of the financial system – and on and on.

Think Obama will be helped when Obamacare goes down the tubes on Thursday? Probably not. Romney is already trying out a narrative that since his health deform plan was blown up the whole term in office is a waste.

Fast and Furious doesn’t seem like a winner for Obama. When they’re doing polls on the president claiming executive privilege over the documents involved in that scandal and you’re on the wrong side of a 56-29 margin, it’s bad for you. Thursday, his attorney general is going to be voted in contempt of Congress – and while they’ll spin that as a partisan GOP attack, the underlying scandal won’t go away. In fact, media attention to Fast and Furious is only going to increase in the short to medium term, and it’s hard to see how that’s going to turn in the president’s favor; to this day it’s hard to understand what innocent and intelligent purpose was served by shipping 2,000 guns to Mexican drug dealers.

Foreign policy? Well, the big foreign policy issues out there don’t seem to be favorable to Obama. In Syria, he has a choice between getting involved in a war that looks like it will be Turkey vs. Iran/Russia/China or he can watch American influence wane – because we’re too late to shape that conflict in a way that would produce a favorable result.

In Egypt, there’s now a Muslim Brotherhood lunatic in charge who can’t wait to do the bidding of the Iranians by starting a war with Israel, while his people attempt a pogrom of the 10 million Coptic Christians in that country and while that country’s capital flight continues in a fashion reminiscent of what happened in Cuba as Castro’s gang advanced on Havana. Considering that the Obama administration cheered the onset of the Arab Spring which produced the very Islamist Egypt we’ve spent billions of dollars over the years attempting to prevent, it’s hard to see how this aids the president’s credibility in foreign policy.

There’s our failed “reset” with the Russians, there’s the ongoing carnage in the Sudan, there’s Venezuela’s aggression against its neighbors, there’s the continuing agitation by Argentina toward the Falklands, which the Obama administration has only encouraged.

In short, unless we’re missing something Obama is running out of ammunition. As Charlie Cook notes, he’s going to be reduced to mudslinging very soon…

We are past the point where Obama can win a referendum election, regardless of whether it is on him or the economy. The success of his campaign is contingent upon two things. First, when focusing on the narrow sliver of undecided voters, between 6 and 8 percent of the electorate, the Obama team must make its candidate the lesser of two evils. It has to make the prospect of a Mitt Romney presidency so unpalatable that about half of those undecided voters will begrudgingly vote for reelection. Polling focusing on the undecided voters reveals they are a deeply pessimistic and angry segment of the electorate and don’t particularly like either candidate (fitting, because they don’t tend to like politicians). But they show signs of being more conservative than not. One unpublished analysis gives Republicans a 10-point advantage on the generic congressional ballot test among those undecided about the presidential race. Close analysis of the numbers shows that Obama might have an edge with between a third and a quarter of the currently undecided bloc. That’s cutting things awfully close.

The second key is turnout. African-Americans look solid for Obama and very likely to vote in high numbers, but young and Latino voters’ turnout appears problematic. Obama’s recent announcement of a newly articulated Dream Act-light policy could help, but it is too soon to see any data showing measurable change. It is what many Latino voters wanted to see, though Obama did it less than five months before the election when it could have been done three years ago. After deportations had reached levels higher than those under George W. Bush, it could take a lot to drive up Latino turnout.

This election is hardly over: The totally unexpected could happen that changes everything. Unless the Obama team can discredit Romney, though, convincing voters that he is a ruthless, uncaring corporate buccaneer, this will be a hard election to win. Probably the only upside for Obama is that the undecided voters appear so sour that they might believe almost anything disparaging said about any politician.

Cook is being charitable toward the president’s chances. His analysis is based on how things look on June 26. When things are moving in a negative direction nationally and voters are cranky, they tend to break against the incumbent – hard. We learned that in 1980. We also learned it in slightly different circumstances in 1968 and 2008.

Romney wins if the election is today, and he’ll win on November 6. The question is by how much. An outspent Obama with no record of accomplishment to run on in a country mired in malaise and declining world prestige is likely to see his hopes only decline in the four months and change to come.

Interested in more national news? We've got you covered! See More National News
Previous Article
Next Article
Join the Conversation - Download the Speakeasy App.

Trending on The Hayride