Update 3: The folks over at Powerline rip Quinnipiac and their mind boggling fictional turnout model:

discount the Quinnipiac results, however, because its sample consists of 35 percent Democrats and 28 percent Republicans. That’s a better split for the Dems than they enjoyed in 2008. There is little reason to believe that the Democrats will have that sort of advantage this year.

The Qunnipiac poll finds that Romney leads among independent voters in Virginia by a staggering 57-36 margin. If Romney actually does enjoy an edge of that magnitude with Virginia independents (and Roanoke College gives him an even larger 26 point edge) it seems almost inconceivable that he will lose the Commonwealth.

Oh pullease. I can tell you the Dems are not going to enjoy the same turnout as in 2008. They did not in 2009 or 2010. If anything, Quinnipiac’s statistical foolery simply confirms the Roanoke College poll results. Obama and the Dems are facing a tsunami backlash like 2010 – except much more focused at The Won!

Update 2: Wow, Dem turnout depressed 70% in FL early voting:

And now we have hard numbers out of Florida showing Democrats well behind their early vote lead when compared to this time last year:

But a Republican yesterday noted that at this point in 2008, Democrats held a 134,774-vote lead in Florida. As of yesterday. Democrats led by less than 41,000 – a nearly 70 percent drop.

The Obama campaign does not dispute those numbers.

Update 1: One polling firm plans to leave its reputation on the floor:

Oh, let’s just skip the rest of the preliminaries and go right to the sample.  The D/R/I on this poll is a ridiculous 45/36/19 that assumes Democrats will add six points to their 2008 turnout while independents largely stay home.  In 2008, recall, the exit polls showed the electorate at 39/31/30, and the 2010 midterm put it at 36/37/28.  Has anyone produced any evidence of such a wave of Democratic enthusiasm?

All those statistical gymnastics to eek out a 50-45% lead for Obama. Who do they think they are kidding? Nat Silver?  – end update

As usual, something happens with polls in the last week before an election to cause them to converge closer to the final result. Is it people finally accepting polling calls? Is it reputations on the line which need to be shored up before the vote so a poll can live to produce another cycle? Who knows. But we are in the final stretch, and the fog is slowly lifting.

Interestingly, both Jay Cost and Sean Trende had posts up in the last 24 hours wondering which picture of the election was going to be more accurate – the state or national polls. Here is Trende on the matter:

Given what we know about how individual states typically lean with respect to the popular vote, a Republican enjoying a one-point lead nationally should expect a three-to-four-point lead in Florida, a two-to-three-point lead in Ohio, and a tie in Iowa. Instead we see Romney ahead by roughly one point in Florida, and down by two in Ohio and Iowa.

…

You can poke holes in this model, to be sure, but I think the simplest explanation is that the state and national polls really are saying different things, at least for now. In other words, if you are calling for the state polls to be right, you are pretty much necessarily calling for the national polls to be wrong, and vice versa.

He added this key observation:

But what if there is a huge surge in Republican voting? Well, this would be felt across all states, at least somewhat, and should be turning the purple states red.

…

For now, I think the best thing to do is wait, and to remember that there is probably more uncertainty in this election than partisans on either side would care to admit.

Jay Cost sees the same thing, applies a different analysis, comes to a similar conclusion:

There is a peculiar divergence between various public opinion polls at the moment. On the one hand, Mitt Romney has built a narrow but durable lead in the national polls, averaging around a 1 percent advantage over the last three weeks. This has cheered the hearts of conservatives everywhere.

Yet, liberals retort, Obama has a lead in enough swing states to add up to 270 electoral votes, and that is really what matters.

What to make of this?

For starters, they cannot both be right.

Cost shows some poll distributions that indicate there are two different turnout models running across pollsters – one group of which is bullish on Democrat turnout meeting or exceeding 2008 levels, the other a little more based in reality.. I don’t see Dem’s increasing their intensity from 2008 , unless those Dems turning out are also switching from Obama to Romney.

Anyway, the real answer is wait and see. And today these gentlemen get a hint at the answer. First in Michigan:

Obama’s lead over Romney has shrunk to just under 3 points, 47.7 percent to 45 percent, with 3.8 percent undecided, according to a new Detroit News/WDIV Local 4 poll of likely voters. Obama’s lead was 6.7 points earlier this month and has eroded to within the poll’s 4 percentage point margin of error. It’s the smallest advantage for the Democratic president during the Michigan campaign.

Obama is, once again, at 47% one week out. Broken record time: for an incumbent US President 47% spells probable defeat. So this blue state has just turned bright purple, meaning this state is trending towards the national polls.

Now onto Pennsylvania:

Less than a week before election day, the Franklin and Marshall College Poll shows Romney now trailing Obama by just four percentage points among likely Pennsylvania voters – 45-percent to 49-percent, compared to a nine point deficit in a September F&M survey.

Blue to Purple.

These states which were once supposedly out of reach are now competitive. This should enhance the pro-Romney turnout on election day even more, now these voters have a chance to send a message to DC. All the while the Obama voters may be sitting back a bit too complacent or ambivalent.

Finally, a crazy poll out of Virginia – where even Hot Air did not want to run with the ‘most likely voter’ result.

Governor Mitt Romney has overtaken President Barack Obama by a very narrow margin in Virginia (49% – 44%), according to a Roanoke College Poll conducted after the Presidential debates.

…

Employing a more stringent screen for likely voters (N=503) increases Romney’s lead to 54 percent to 41 percent …

54-41% in Purple VA? Well, if Sean Trende needs to see a purple state turn red, here is my home state as evidence.

I am mot at all surprised over this result. The year after Obama won the 2008 election and took Virginia, buyers’ remorse had settled into the Old Dominion. That year, 2009, Virginia went to the polls to elect a new governor. Up until the Obama win the state had been turning bluer and bluer, with 8 years of Democrat Governors previously. That year, Republican Bob McDonnell won the election 58-41%.

McDonnell’s 2009 numbers look very similar to Romney’s in this poll. So I don’t think it is far fetched at all.

Also note the coat tails:

Employing a more stringent screen for likely voters (N=503) increases … [GOP Senate Candidate George} Allen’s lead to 51 percent to 39 percent. …

I am still bullish on this election. And I just don’t think it will be the nail biter everyone claims it is.

Tags: 2012 elections, MI, PA, Polls, VA