Considering the assault Obama has been under, he's holding up.

I think it's a good idea to continue to spend in Florida and keep up the heat there, McCain may not be spending money there now, but eventually they will taking resources away from other states.   Obama does not need Florida to win, but I think McCain needs Florida for him to win.

I agree that Ohio is statistically tied, with a swing of Obama +2 to McCain +1.

Florida, however, is of much more concern, and I think it's wrong to say that these results are "within the margin of error."  It's gone from Obama +2 to McCain +4, a swing of 6 points.  That's significant.

I'm not comfortable being defeatist about Florida, suggesting that it's not a necessary part of our electoral landscape.  It's a huge factor in the election, and I want to be sure we are always contending there.

Some delusional person in MyDD colored Oklahoma blue on the map. You can't be serious!

Pennsylvania is to Mccain , what Florida is to Obama , both would eventually lose both states...

Is that in the national tracking polls, Obama seems to have stalled and McCain has inched ahead since early August. If Bradley effect is   a reality and not a figment of imagination, then November will be really an uphill climb. Maybe after the conventions, Obama will edge ahead; but I am thinking that the election is likely to be decided in the debates.

Florida is the ultimate battle ground state for Obama.  The question that needs to be answered here is if the McCain campaigns choice to not spend money advertising is telling.  Are they so confident in there chances there that they no ultimately there is no way Obama can win the state due to most likely his inability to connect with Seniors and the overall demographics.  Looking at Ohio this state is moving more and more to the democratic column.  Recently with the success of Strickland and Brown, Ohio offers more hope for Obama.  However, race still seems to be an issue with Obama.  In order for Obama to win there he MUST win the Cleveland area by a 80% margin along with a strong showing in Columbus and the other surrounding College towns.  Pennsylvania is in the bag for Obama due to the democratic registration advantage that democrats possess there.

McCain's latest Michigan numbers weren't good for him either.   If he can't win either MI or PA then he is done, I think.  Because I do think Obama, and Democrats as a whole, will win in at least 2 states that Bush won in '04.

Good points by all. I still think though, that if we can pick up Ohio and the Colorado Coalition, we'll be all right.
Pennslyvania is not a battleground state and hasn't been for a long time.  And if it wasn't before Biden, it sure isn't now.

Florida was never going Democratic against John McCain with  Joe Liberman campiagning there. Not with Hillary and defintely not Obama. The best we can hope to do is a tactic designed to get McCain  to spend more money there.  McCain and Libermann will tie up the veteran, Jewish and Senior votes. What else is there?
 

If Obama picks up Ohio he'll win.  Same with Virginia, Colorado, Florida, and Nevada/New Hampshire.  Of the Kerry states, the only one where he seems vulnerable is New Hampshire (and I think he'll win there).

Here's the good news.  Polls seem to show that all five states/combinations are winnable, and victory in each probably turns on a distinct part of the electorate.

Eg. the defection of rust belt Dems in Ohio probably wouldn't hurt him in these other areas, because they're not part of the rust belt.  Latino turnout in Nevada and Colorado probably isn't connected to what we think of as the mainstream race (that's a different group with their own issues).  Heavy African-American turnout could save Obama in two of these states (but would do nothing in the other four).

The bad news, of course, is that one could imagine Obama's losing all of them (which wouldn't doom him necessarily, there are still some odd combos which could put him over the top, Indiana, for instance, or Montana/Alaska).  And, of course, it could go the other way (he could win all of these states/combos, but for different reasons).

One odd fact which seems to be showing up in the Ohio and Florida polls.  Dems in Ohio seem to be much more upset with Obama than Dems in Florida (talking about the rate of defections and fence sitting), which isn't what most people would have predicted, I suspect, last June.

One explanation is that Clinton's determined campaign in Ohio put a lot of people off Obama, but Florida voters didn't really care all that much about what happened during the primaries, and we're seeing the effects of both.

Call it a gut feeling, but I think Obama will win Ohio by 1 point, and with it the election (which doesn't mean he'll lose these other combos, but he might).  Why?  Because the Dems control Ohio's state government now, and are making it much easier for Dems to vote (no repeat of what happened in 2004).

Some comments on the polls

1. I am troubled that the decline in Obama's poll numbers in the battleground states has happened at the same time that the Obama field organization has been in full swing opening offices, recruiting volunteers and registering new voters.  Are McCain's television adverts trumping the combination of the Obama adverts and the field organization?

2. I always suspected (but have no proof) that many of the newly registered Democrats in the northern industrial states (new voters and party switchers) voted for Clinton in the primaries in order to register their disapproval of Obama.  If so, we are not seeing Hillary die-hards holding out, but we are seeing anti-Obama voters moving from one anti-Obama candidate to another.  Nothing much to do with Hillary.

3. Do we know anything about how enthusiastic Obama and McCain supporters currently are?  We never have a near 100% turn-out.  Maybe the strategy is that a lot more McCain supporters will not vote, as opposed to Obama supporters who will vote.  (Although I presume that the Rove 72 hour drive will be in effect).  So, McCain could turn out to be the more popular choice for president (among citizens but not 2008 GE actual voters), but Obama still might win on technical points by having a more enthusiastic group of supporters who actually vote.