New Swing State Polls Show Romney Falling With More Trouble Possibly Looming

We’ve got several new swing state polls in over the past couple of days as well as two polls on the reaction to Romney’s 47% comments. The data, on the whole, points to Romney falling further behind in the key swing states — Ohio, Virginia, and Florida — with the potential to fall more as the 47% comments sink in with voters. FOX News dropped some numbers today — Obama up 7% in Ohio, 7% in Virginia, and 5% in Florida — that will likely cause the Romney campaign to get more panicky in the coming days, and further increase the already growing sense of doom surrounding Romney.

First the numbers, then my analysis.

* OHIO: Obama 49%, Romney 42% (FOX News)
* FLORIDA: Obama 49%, Romney 44% (FOX News)
* VIRGINIA: Obama 50%, Romney 43% (FOX News)
* VIRGINIA: Obama 52%, Romney 44% (Washington Post)
* VIRGINIA: Obama 50%, Romney 46% (CBS/NYT)
* WISCONSIN: Obama 54%, Romney 40% (Marquette University)
* WISCONSIN: Obama 50%, Romney 46% (CBS/NYT)
* COLORADO: Obama 48%, Romney 47% (CBS/NYT)
* COLORADO: Romney 47%, Obama 45% (Rasmussen)
* NEW HAMPSHIRE: Romney 48%, Obama 45% (Rasmussen)

Clearly most of these numbers are very bad for Romney, particularly the Florida, Ohio, and Virginia numbers.

One thing that is clear — Wisconsin is not as close as it seemed to be a few weeks ago. We now have two new Wisconsin polls, and the average Obama lead across those two is 10%. In my article earlier today about six things to give the Romney campaign hope, the possibility of flipping Wisconsin was one of them. Without Wisconsin, Romney’s options of getting to 270 become very narrow. If he loses Wisconsin, he could win Ohio, Florida, AND Virginia and STILL lose the election (and he is currently trailing all four of these states).

The FOX News numbers in Ohio and Florida, where he is behind by 7% and 5% respectively, is a bit on the scary side for Romney. These numbers are more than what Obama won the states by in 2008. If Romney cannot outperform John McCain in these states, it raises serious questions about his viability in the other swing states.

And Virginia is also pumping out mostly bad numbers for Romney (he’s now trailed in seven straight Virginia polls) — particularly the Washington Post poll which has him down by 8% and the FOX News poll which has him down 7%. Again, Obama is outperforming his 2008 margin in Virginia according to these polls. Virginia, Ohio, and Florida are all must wins for Romney, and he is trailing outside the margin of error in all three, according to virtually every poll except Rasmussen polls.

Romney is holding up in Colorado, which looks like a total toss-up at this point, and got a good number in New Hampshire. But those states don’t matter if he continues to fall behind in Virginia, which appears to be the case, and remains behind in Ohio (he’s trailed in five straight Ohio polls). Colorado and New Hampshire are unlikely to be states that are tipping point states (states that will put a candidate over 270 electoral votes). Nate Silver’s model at 538 lists the top tipping point states as Ohio, Florida, Virginia, and Wisconsin. Romney is trailing in all of them. And obviously if Romney is behind in Florida, the talk of Romney winning disappears and the talk of another Obama electoral college landslide begins.

So what is causing Rasmussen to have what appear to be the only favorable numbers for Romney (and by favorable, I mean he is simply losing by less — even Rasmussen has him trailing in Virginia, Ohio, and Florida), while all other swing state pollsters seem to be showing Obama pulling away? Part of it is the modeling. Pollsters are projecting vastly different turnout models. Rasmussen has the most Republican-friendly model, while some other pollsters have more Democrat-friendly models. What does that mean? Turnout will be crucial, and pollsters all seem to disagree on who will turnout and in what numbers. In 2008, 7% more Democrats voted than Republicans. Will that be 6% this year? 4%? Or will it actually be more Republicans voting this year than Democrats, which is what Rasmussen’s polling model projects? It’s difficult to say, and we won’t know for sure until the votes are in. What we do know is that Democrats have a large advantage among registered voters. What we don’t know is how many of those registered voters will simply not vote this year.

The other part, which Nate Silver talks about in detail here, is which pollsters poll cell phones, and which don’t. Pollsters who have been polling cell phones — NBC/WSJ, CBS/NYT, etc. — have shown much stronger results for Obama. Pollsters who do not poll cell phones — Rasmussen does not — have shown much better results for Romney. Nearly 1/3rd of Americans now rely exclusively on cell phones and do not have landlines, meaning a poll done by Rasmussen would exclude them entirely from the sample. The types of people who only use cell phones tend to be younger, live in big cities, and be more Democratic-leaning.

In other words, if Romney is hoping that all pollsters showing him falling further behind are wrong, and only Rasmussen is right in that he is still in virtual ties, that seems to be grasping at straws.

Then there is the other new variable — none of these polls have fully accounted for the fallout to Romney’s 47% comments yet. We won’t really know what impact that story has had — if any — until probably next week. But we do have two polls in measuring the initial reaction of voters to the 47% comments:

* Gallup has a new poll up finding that 36% of voters said that Romney’s 47% comments make them less likely to vote for him, 20% said they made it more likely they’ll vote for him, and 43% said the comments will not impact who they vote for. Among Independents, 29% said the comments make them less likely to vote for Romney, 15% said more likely, and the rest said no difference.

* Reuters has polled the topic and found that 43% of voters say Romney’s comments make them less likely to support him. 59% said that Romney unfairly dismissed half the country.

How that is all reflected in state polls remains to be seen, but if the above two polls are correct, Romney’s numbers could get worse next week.

And then there are the Senate races, which have taken a big shift to Democrats since the conventions. Republicans were favored to take over the Senate majority just a few weeks ago, now the Democrats are favored to retain the majority. Key races, like Claire McCaskill taking a lead over Todd Akin in Missouri, Ben Nelson taking a lead over Connie Mack in Florida, Tim Kaine taking the lead over George Allen in Virginia, Elizabeth Warren surpassing Scott Brown in Massachusetts, and Sherrod Brown holding on to a fairly comfortable lead over Josh Mandel in Ohio all seem to point to a national shift to Democratic Senate candidates. National shifts do not happen in a vacuum, they usually happen for reasons that benefit all the way up and down the ticket. That would be a good sign for the top of the ticket — Obama.

Couple the bad Senate polling for Republicans with the fact that several Republicans, some in House and Senate races, have now spoken out against Romney’s 47% comments, a sign that their internal polling shows the comments as potentially damaging, and all of this seems to point to potentially larger problems on the horizon for Republicans in House and Senate races.

All that being said, I have to continue to reiterate a basic fact. In the current hyperpartisan era, even major news stories often only impact polls slightly, and with seven weeks to go, a lot can and probably will change.