I seriously doubt if you continue to work to register and turn out Northern Virginia that 47% of VA voters really have a negative opinion of Obama.

It's cause for "concern" (hate that word), but that number is certainly fungible.

http://strategy08.wordpress.com

Virginia will be very close but I don't think DEMS should bet the election on this state. Virginia is still a southern state....that is trending DEM. Maybe Warner's anticipated landslide can help sweep Obama to a narrow win.

I see Missouri, Indiana, Iowa, Colorado, New Mexico and Nevada as far more likely pickups. The Midwest and Mountain West/Southwest seem to be the most fertile grounds for DEM gains this year.

Indeed, the bad news just keeps piling up for Obama.

Hopefully he will see the light and announce that he is dropping out today.

I don't mean to sound like a dick, but I'll never understand the comments of the people who say it's not worth a shot or only a limited effort. Even if living there, as opposed to New York where I live, gives you some insight into voting habits around you, the Obama campaign is run by professionals who have made several excellent calls so far. They know how to look at recent history but also the base numbers and the potential they possess. They clearly see an opening, so why not let them go at it?

If come the middle of September, they have information that they aren't making a difference, they can always scale back their efforts. That's what every campaign does. But why not work at it now? The public polls show a very close race. Why should internal polls be any different?

We always hear talk of a 50-state strategy. Well, this is what it looks like, at least as far as first steps go. Even if we end up losing this time, we are registering voters and collecting other information that will give us a leg up next time. That way, when Obama is running for re-election in 2012, or some other Democrat is running, this person will have a base in the state. Did John Kerry really give us that from 2004?

Also, I don't think anyone is predicting that this will be the next Rhode Island. No, it'll probably be more like the next Pennsylvania or Wisconsin, a usually competitive state that always goes a few points in one direction or another.

If you have a generally favorable political environment and tons of money, it makes all the sense in the world to expand the playing field even if you don't think you can win all of those states. It forces the opponent to spend time and money in those states and to give up on playing offense in states like Wisconsin or Michigan, thereby freeing your resources in those states. Yes, this strategy has its own risks, but still, somewhat counterintuitively, it is still safer than just targeting narrowly your best states that put you barely over 270 EVs, because then it is enough for one thing to go wrong to cause you to lose the election.

Eric, you're making a big statistical mistake in buying Rasmussen's narrative of the internals showing "concern" for Obama.

The concern in this poll is actually for McCain because of the turnout model. Rasmussen's crosstabs don't provide the turnout model, but that can be extrapolated from the data they DO provide in the form of subgroup preferences. By my math, this Rasmussen Virginia poll had to have limited the black vote to no more than 14% and the white vote at no less than 81%. In 2004 exit polling showed the black vote at 21% and the white vote at 72%, and in the 2006 midterms the exit polling had the black vote at 16% and the white vote at 78%.

In this poll, Obama pulls 38% of whites, 95% of blacks, and 80% of "not sure" which frankly I take to mean other nonwhites in what must have been poor question word choice. These are big improvements over Kerry in all three categories; Kerry got 32% of whites, 87% of blacks, and 61% of other nonwhites. If you apply the last presidential election turnout model to this Rasmussen Virginia poll, these subgroup vote shares are enough to put Obama at 53% of the total vote. If you reduce Obama's share of "other nonwhites" to the same 61% that Kerry got, Obama still easily clears 50%. If you apply 2006 turnout, Obama falls just a hair under 50%, at 49.64.

Finally, regarding Obama's "high unfavorables," his favorable is at 52-47, McCain's is at 64-36, and yet the 12-point favorability deficit translates to a measley one-point 48-47 trial heat deficit. This just reinforces that this election is all about Obama: almost everyone who likes Obama votes for him, whether or not they like McCain, too. And an election that is a referendum on Obama is one where 52% favorability is enough to have an even chance to win. Sure it would be nice to have it higher, and Obama needs to get it higher. But it may actually BE higher among the actual electorate based simply on including substantially more black voters than Rasmussen polled.

My wife and I when we travel conduct our own poll of voter preferences by tracking the number of bumper stickers and logging which candidate they are for. We traveled last weekend in NC for 6 six hours and the totals were: Obama 254 - McCain 1...

I don't know about you but that tells us that NC is in play for Obama and we are this far out from the election. Anybody seeing alot of McCain bumper stickers?