Why an Electoral Landslide Could be Headed Obama's Way

As the election gets closer - it's barely a week away - pollsters and pundits may still be agonizing over what happens on the day of November 6, but the election is not actually on November 6. Rather, the election ends on November 6. The election is now. Millions of Americans are casting early ballots, and in record setting numbers that seem to already be blowing past the previous records set in 2008. Early voting was the President's strength in 2008, and it will certainly be that way again this year. Just some numbers coming in from across the country and from crucial states:

Florida, which just started voting on Saturday, reports numbers already above records:
The Times' Adam Smith reported on Twitter, "More than 20k pple voted eary today so far in Hillsborough Co...In 08 biggest single day (11/1/08) was 18,736.

The Miami Herald's Marc Caputo tweeted, "Miami-Dade early voting so heavy that by 3 pm 14,745 people voted -- more than ENTIRE first day of 08 EV (12,000). 12-hour total: 22,625."

Gary Fineout of the Associated Press, "Leon County - a Democratic stronghold in Fla - had a record turnout for early voting with 5447 votes cast on day 1."
And among absentee ballots returned thus far, the President's campaign has cut the GOP advantage from 16 points to a mere 5 points this time. Remember that the president won Florida in 2008 despite that enormous GOP advantage in absentee voting.

In Ohio, the President may have already mounted a lead in early voting that will be difficult, if not impossible, to overcome on election day. According to the recent TIME poll, the President is clobbering Mittens 2-to-1:
Obama has clearly received a boost from Ohio’s early voting period, which began on Oct. 2 and runs through November 5. Among respondents who say they have already voted, Obama holds a two-to-one lead over Romney, 60% to 30%.
The poll found that 20% of voters already voted - if that's true, then the president already has a 6-point advantage going into the final week of the campaign. Even polls in Ohio that purportedly show a dead-even race overall show a similar advantage for the president in early voting.

In a pretty stunning, but under-noticed development, Obama even leads the early voters in North Carolina - a state pundits have written off in favor or Romney. And the president leads in early voting in North Carolina in a poll conducted by none other than the conservative pollster Rasmussen, even as they found an overall lead for Romney:
The race remains unchanged from a week ago, so North Carolina remains Leans Romney in the Rasmussen Reports Electoral College projections.  In 2008, Obama was the first Democratic presidential candidate to carry North Carolina in over 30 years.

However, the president leads 52% to 46% among the 35% in North Carolina who have already voted.
If those numbers are correct - let's assume for a moment that Rasmussen isn't right-skewed - that gives the president a 2.1-percentage point lead in total vote in North Carolina heading into the final week of the campaign. And that is the Republican poll's numbers. Public Policy Polling's numbers from last week has the overall race tied, and the President up a whopping 57% to 42% among early voters. And according to the Obama campaign, Democrats outnumber Republicans 50% to 31% in NC early voting. Obama won North Carolina by a razor-thin 0.35 points in 2008. No wonder David Axelrod is making clear that the Obama campaign in North Carolina is in it to win the state.

Axelrod is also saying that the campaign is building up a large lead in Iowa:
“We believe that we are mounting up a very, very large lead in Iowa based on where those early votes are coming from,” he said, while indications in Florida “are very positive.”
Campaign Manager Jim Messina expands on why these early voting numbers are so crucial:
“What early vote does is help us get out our low propensity voters -- voters called sporadic voters -- which broadens our universe and frees up more ‘get out the vote’ resources later, especially on election day,” Messina said. “Our numbers and public numbers are showing that more Obama sporadic voters are voting than Romney sporadic voters, which is a very big piece of business for the total turnout.”
Messina is saying something very telling here. Think about this: who are the low propensity voters? Voters often not counted by pollsters in their 'likely voter' models. More of the president's voters are low propensity voters, proportionately, than Mr. Romney's voters. This isn't a surprise; conservatives tend to vote more consistently than liberals (liberals, are you listening?), and older Americans (aka Romney's strongest block) vote more consistently than younger Americans (aka Obama's strongest block).

The 'likely voter' models are banking on the assumption that there is something like a 7-10 point swing in Romney's direction when going from a sample of all registered voters to likely voters. This weekend's Reuters/Ipsos poll shows Obama narrowly leading nationally 49-46, but among registered voters, the president leads by a whopping 12 points, 51-39!

But what does that mean in real terms? It means that any increase in Republican early voting probably comes out of their numbers on election day, while a good amount of the increase in Democratic early voting probably comes from voters who wouldn't otherwise vote at all. The Democratic voting universe is far more expandable than the Republican one. If it weren't, Republicans wouldn't be trying so hard to stop people from voting.

Incidentally, the Princeton Election Consortium, which is based only on state polls, at the moment has the president sitting comfortably at a projected 300 electoral votes, with its probability models indicating more than 90% of possible outcomes resulting an Obama victory:


The early voting surge is growing even in states that aren't swing states, which could point to the president not only building a strong electoral college victory but a sizable popular vote margin as well. Nationwide, the president is leading Mitt Romney by 15 points among voters who have already cast their ballots. If something like this lead holds, and given the estimate that as many as 40% of voters may cast ballots early, little can be done to prevent an Obama landslide, regardless of how many angry old white men Mitt Romney can bring to the polls on election day.

One of the national pollsters out there with an innovative approach to polling has been the RAND corporation's American Life Panel - which provides both a rolling average and combines the probability that someone will vote with the probability they will vote for a given candidate. Here's where the RAND poll was as of Saturday:

Lines falling outside the gray area indicates a difference not due to sampling error. If you look at it carefully, the president's lead now is as large as it has ever been since the beginning of this track.

So let's put it together: early voting is surging everywhere, and especially in the battleground states. This early voting overwhelmingly favors the president, more of our early voters are unlikely voters and the only early voting category that Republicans have traditionally had an advantage in - absentee voting - has seen that advantage cut significantly. Nationally, the president is not just leading Mitt Romney by double digits in early voting but in the only national poll that has had a constant sample (i.e. it didn't suddenly move to a more southern, white sample), the president's lead seems to be expanding, in contrast with the horse-race national polling by media outlets.

What do all these translate into? Well, nothing, if we don't keep working as hard as we can over the next 8 days to re-elect this president. But if we do, if we get this president's back as he has always had ours, if we vote, if we get the vote out for him, if we give it everything we've got, an Obama landslide is more than possible.