What if Gallup is right?

I'm a "science" guy. I tend to accept polls as being the distilled magic of statistics. The battleground state polls have shown Obama leading throughout this campaign, so, all things being equal, I rate Obama the favorite.

On the other hand, the polls are screwy. On the third hand, there is typically a word for people who spend their time complaining about the polls - "losers."

On yet another hand, there is this perspective offered by Bob Krumm:

Meanwhile, in what should have sent up alarm bells across the land, Barack Obama’s campaign was diminished to arguing that it would win on the strength of its get-out-the-vote efforts with youth and minorities. Second to mentioning Harry Truman, this is the last refuge of the losing side. Almost never is depending on sporadic voters a recipe for success. Even in 2008, Obama’s leads with those demographics only padded the margin he had already won because of his eight point lead with independents. Falling precipitously from his earlier indy numbers, he and his acolytes should have known that calamity lay ahead.

Most pollsters took a beating in predicting the race. But that should have been expected too. The last two times that a Republican challenged a Democratic incumbent (1996 and 1980) the polls overestimated Democratic support by 5.1 and 7.2 points. And ‘96 was not even in bad economic times.

The biggest gap was not one of gender, but between the opinions of those who voted and those who did not. Those most likely to vote–whites, property owners, investor class, church-goers, married, and the elderly–despised where Obama had taken the country by margins of sometimes more than 20 points. Singles, minorities, and the youth still backed the President, but with diminished support and turnout from four years before.

The entire outcome hinged on who turned out to vote, and when only nine percent of the electorate wished to talk to pollsters, projecting who that was likely to be, became a fool’s errand. But statisticians are not above being fooled. A look at any one of those demographics–even the ones that still supported him–would have showed the President’s downward shift from four years before. But it was too easy to weight bad samples toward a guesstimate of a turnout model instead. Only Gallup got it right, and the result was nasty for Democrats. They went from a 12-point advantage in party identification among likely voters in 2008, to a one-point disadvantage in just four years.

That may be the model that is informing George Will and Michael Barone's predictions.

Complete polling nonsense.

CNN says it's all tied in a D+11 poll.

That's way better than 2008.

Romney up 1 in Michigan.

The poll was done by a Democrat polling firm.

A new poll in Michigan has Mitt Romney leading by 1 point in a state that’s considered leaning toward President Obama.

The poll, conducted by Democratic polling firm Baydoun-Foster for Fox 2 News Detroit, has Romney up 47-46 on Obama. The same polling firm pegged the race in Michigan a tie in a survey late October.

The poll is the first in Michigan since August to give Romney a lead, however, while Obama holds a 3.8 percent lead in the Real Clear Politics polling average.

Baydoun-Foster’s Michigan polling has leaned more toward Romney than other Michigan surveys by several points throughout the final months of the campaign, a review of RCP’s Michigan polling shows.

More Acid Indigestion for Chicago.

Michigan is all tied up - Romney 46% to Obama 46%.

Acid indigestion in Chicago.

This was not part of Obama's plan.

Tribune poll says Pennsylvania is too close to call.

President Obama and Republican Mitt Romney entered the final days of the presidential race tied in a state that the campaigns only recently began contesting, a Tribune-Review poll shows.

The poll showed the race for Pennsylvania’s 20 electoral votes locked up at 47 percent in its final week. Romney was scheduled to campaign in the Philadelphia area on Sunday, and former President Bill Clinton planned to stump for Obama on Monday. The campaigns have begun to saturate the airwaves with millions of dollars in presidential advertising.

If Pennsylvania is in play, and it seems to be, then that implies that there really might be a wave lifting Romney throughout the country.

Poll puts Romney up 1 in Minnesota.

Via left-leaning Politico:

Barack Obama and Mitt Romney are separated by just 1 point in Minnesota, effectively making the race there a toss-up, according to polling taken for the conservative American Future Fund.

Romney takes 46 percent of the vote to Obama's 45 percent in the poll, which was conducted by the GOP firm NMB Research and shared with POLITICO. The Republican presidential nominee is up 13 points among independents, ahead of Obama 49 percent to 36 percent.

They survey breaks with recent public data, which has shown Obama maintaining a single-digit edge over Romney, but gives Republicans reason to hope for an upset. Democrats aren't taking the state for granted at this point, with Obama countering pro-Romney ads (including AFF advertising) and deploying Bill Clinton to campaign there.

In a polling memo, pollster Glen Bolger attributes the closeness of the race to Minnesota's overwhelmingly white population.

"Minnesota is very much a battleground state due the low minority population of the state and President Obama’s problems with white voters. Romney has a good chance to pull off one of the biggest upsets of the election cycle in this state," Bolger writes.

The poll tested 500 likely voters on Oct. 30 and 31 with a margin of error of plus or minus 4.38 percentage points.

Pennsylvania's Most Accurate Pollster says Romney up by 4...

...in Pennsylvania.

Mitt Romney and Republican allies have suddenly poured more than $10 million into Pennsylvania to fund a late push by the presidential challenger. Two weeks ago, a poll by Susquehanna Polling and Research showed Mitt Romney leading Barack Obama by four points in the Keystone State, 49-45%. They are set to release the results of a new poll Sunday morning. Will they point t o a Romney win? (Note: a previous edition of this article incorrectly stated that the 49-45% Romney lead was this weeks poll).

In what could prove to be a brilliant campaign move, the Romney campaign hid details of a Sunday campaign appearance in Pennsylvania until after the Obama team released their final campaign schedule. Critics have been dismissing Susquehanna's tight polling for over a month based on little more than the fact that a Republican presidential candidate hasn't won Pennsylvania since the 1980s. George W. Bush came close, however, in both of his campaigns. Republicans also took the US Senate seat and Governor's mansion in 2010, despite Pat Toomey being called "unelectable" by the media.

If Romney wins Pennsylvania, the race is over.

It is time to stop paying attention to the RCP polling average.

The RCP shows the race narrowing because it includes a Congressional Connection poll showing Obama up by 5 points.

In its likely-voter model, the Congressional Connection Poll projected that the 2012 electorate will be virtually unchanged from 2008, with Democrats holding an 8 percentage-point advantage among voters (compared with 7 points last time) and whites representing 73 percent of voters (compared to 74 percent last time).

So, this poll assumes that Obama will have the same level of enthusiasm that he had in 2008, despite every other poll showing that Romney is winning among independents?

This poll is designed to do one thing: prevent the Dems from panicking.

Also, the RCP cited the Battleground Poll as a +1 Obama, even though the polls itself explained that it was +5 Romney based on likely voters.

Update: This PJ Tatler pieces demonstrates that the Congressional Connection poll actually weights the "likely voter" responses to favor the Democrats.

So, basically, last month when the polls favored Obama, we were told that partisan identification was not something that pollsters weighted, that they were just telling us the information they were being provided. Now, this is the second poll where the pollster has clearly upped Democrats (from 35.6% to 37%) and lowered the Republicans (from 31% to 29%)

This is "polling as propaganda."

CBS Poll finds Obama has a 1% edge...

...but the news gets worse for Obama thereafter.

First, it is a poll based on a "weighted" 563 likely voters.

Second, the weighted partisan breakdown is 36% Democrat, 31% Republican and 33% Independent. Since Republicans have a 1% edge in registration and more motivation this weighting is optimistic for Obama.

What is also interesting is that the CBS internals give the lie to the idea that the polls are just "calling what they find." In order to get to those probably distorted partisan weights, it seems that CBS had to reduce the number of Republicans it actually found from 257 to 242 and increase the Democrats from 297 to 321 (in the registered column.) CBS doesn't tell us the unweighted numbers for the "likely voters" for some reason.

GALLUP: Romney Up 52-45% Among Early Voters...

...according to Gallup.

Rasmussen: Romney up two in Ohio.

Here's the Rasmussen Report:

The race for Ohio’s Electoral College votes remains very close, but now Mitt Romney now has a two-point advantage.

The latest Rasmussen Reports telephone survey of Likely Ohio Voters shows Romney with 50% support to President Obama’s 48%. One percent (1%) likes some other candidate, while another one percent (1%) remains undecided. (To see survey question wording, click here.)

Ohio remains a Toss-Up in the Rasmussen Reports Electoral College Projections. Based on the current projections, Romney would have to win Wisconsin if he loses Ohio in order to move into the White House.

The candidates have been locked in a very tight battle in Ohio since August. A week ago, Romney and Obama were tied in the Buckeye State with 48% support each. This is the first time Romney has taken even a modest lead in the race.

Nearly one-in-three Ohio voters (32%) have already cast their ballots. Obama leads 62% to 36% among these voters. Romney has a large lead among those who still plan to vote. The question of who wins Ohio may come down to whether enough Romney voters get to the polls on Election Day to overcome the president’s lead among early voters.

Washington Post proves that Limbaugh was right.

The current Washington Post poll has it 50 to 47 in Romney's favor.

The interesting thing is that the Democrat-Republican difference in the number polled was 4%, i.e., 34% Democrat to 30% Republican - not far off from the predicted turn-out for this year, which has been predicted all summer!

On the other hand, on 10/12/12 - two weeks ago - the Washington Post's +3% for Obama had a 9% spread in favor of Democrats - 34% Democrat to 25% Republican.

So, was it the case, as Limbaugh predicted, that all of the polls up to the last day or so were simply exercises in depressing the Republican base but now that the election is looming, the pollsters have to get "serious" in order to keep their reputation intact for next time?

Makes you wonder.

Another poll for Romney outside the margin of error - Romney 48% - Obama 45%.

From Monmouth.

That puts three polls in Romney's favor outside the MOE.

Rasmussen has it Romney 50% - Obama 46%...

...outside the margin of error and in the neighborhood of Gallup's 6% edge for Romney.

Does the demeanor of Romeny and Obama tell us what the polls are really saying.

Jonah Goldberg writes:

Well, it looks like Romney took my advice (no not literally). In boxing terms, Romney went for a clinch. While Obama constantly tried to push Romney off to start a real fight and heighten the contrast. In other words, Obama behaved like a challenger in trouble and Romney acted liked a candidate with something to lose. That tells you something about the poll numbers both of these guys are looking at.

And while I obviously think Romney followed the right strategy, I have to say I wish he’d broken out of the clinch a couple times to land a some easy blows. It was often painful to watch Romney blur the difference between them. Still, I think he was smart not to take the bait.

Meanwhile, Obama looked a bit desperate at times, particularly when he was staring at Romney as if to force the former governor into spontaneous human combustion.

Ultimately, I think this debate served Romney well, even if the flash polls say Obama won narrowly on points (the polls said the same after the second debate and Romney’s numbers continued to improve). If you went into this debate worried that Romney isn’t a safe presidential pick, you came out reassured. Also, the long tail of this debate will play well to Romney’s advantage in that he had a few more memorable lines and was better on the economy. Obama’s one really memorable line — about the navy and bayonets — will not help him in swing states or with constituencies not already in his column.

Post Debate 2 Gallup has Romney at +7.

Gallup: Romney 52% - Obama 47% among likely voters.

Preference Cascade, anyone?

He's not "defying gravity" anymore.

Romney up 6 among likely voters...

...according to Gallup.

The polls this year have been all over the map. Washington Post-ABC had Romney down 3 (with a +9 Democrat advantage), which means that there is a 9 point discrepancy.

But, you know, these polls are Science!, so who are we to question them.

More Weird Polling.

The Washington Post's recent poll giving Obama 4 points on Romney makes Romney's RCP advantage disappear, but can the WaPo poll be taken as anything other than designed to make it appear that Obama has regained ground before the second debate. The Weekly Spectator reports:

The newly released Washington Post/ABC News poll of likely voters says that if the election were held today, Democrats would enjoy a 9-point advantage over Republicans in voter turnout (35 to 26 percent), and President Obama would beat Mitt Romney by 3 percentage points (49 to 46 percent). The good news for Romney is that, to be able to cut a 9-point disadvantage in voter turnout to just a 3-point deficit in votes, he must be doing awfully well with independents. The bad news for Obama is that it’s very hard to find corroborating evidence of this massive Democratic edge.

So, if we believe the pollsters, more people are identifying themselves as Democrat, even though they don't like Obama?

Does that make any sense?

Politics and Polling.

Yes, Virginia, politics does influence polling. For example, Gallup changed its polling methodology to favor Democrats in response to White House criticism.

More Polling Weirdness - Major Polling Operation Declares Florida, Virginia and North Carolina are in the Romney column.

Here is the story:

In a surprise move, a major political polling company is pulling out of Virginia, declaring that President Obama cannot win the state.

“I think in places like North Carolina, Virginia and Florida, we’ve already painted those red, we’re not polling any of those states again,” Suffolk University polling director David Paleologos said Tuesday night on Fox News’ ”The O’Reilly Factor.”

“We’re focusing on the remaining states.”

Suffolk’s most recent Virginia poll, issued Sept. 27, showed Obama with a 46-44 lead over GOP challenger Mitt Romney, with seven percent undecided.

Paleologos, noting that Obama’s lead was within the statistical margin of error, said at the time that the president maintained “a decided popularity advantage.” Obama’s “favorable” rating was 52 percent, compared with 42 percent for Romney.

“However, with job approval and head-to-head numbers stuck at 46 percent, it will be a significant challenge for Obama to convince the remaining undecided voters to re-elect him,” Paleologos cautioned.

Obama also had maintained slight leads in Suffolk polls in Florida and North Carolina.

The Boston-based polling center has conducted surveys since 2002 in states with what it claims to be “a 96 percent record of accuracy.”

This may be true, but isn't this more akin to "reading the tea leaves" than science?