Polls Tighten Just a Bit, Electoral Map Stable, A Home State Betrayal?

PROJECTED ELECTORAL VOTE: OBAMA/BIDEN 375, MCCAIN/PALIN 163
(no changes)
PROJECTED POPULAR VOTE: OBAMA/BIDEN 52.1%, MCCAIN/PALIN 45.9%
(Obama/Biden down 0.4%, McCain/Palin up 0.4%)
TIME UNTIL FIRST POLLS OPEN: 7 DAYS, 9 HOURS

Well, I said yesterday that McCain would need to close in by 0.8% per day for the last 9 days of the race to be even come election time. What do you know? McCain closes by 0.8% today in our projection. Whether he can sustain that momentum every day for the next 8 days remains to be seen.

We got a slew of new polls today (28 in all!) but they resolve nothing as the electoral count remains the same. I'd said I'd love a newer poll in Arizona and I got one, and you see why I wanted one. The polling data in North Dakota and Montana still badly needs updating.

McCain now needs to pick up 13 electoral votes (a state the size of Virginia) per day the rest of the way in order to get to 270 on November 4th.

The state polling may not have fully caught up with the national polls as the aggregated state projections would point to a 7.2% win for Obama, whereas the national tracking would point to a 6.2% win for Obama. If you assume a possible 1 point swing, it would move Missouri back to McCain's column, but no other state at this point.

STATE CHANGES
None

KEY BATTLEGROUNDS
Indiana, North Carolina and Missouri all remain as key battlegrounds. Missouri is about as close as they come right now.

SERIOUS BATTLEGROUNDS
Nevada, Florida and Montana all remain as key battlegrounds.

SUBSTANTIAL BATTLEGROUNDS
Arizona -- surprise, surprise, John McCain's home state is now a battleground. I don't seriously expect him to lose here, but it has gotten much closer here.
Georgia also remains as a substantial battleground

FRINGE BATTLEGROUNDS
Ohio, Colorado, Virginia, New Hampshire and North Dakota all remain as fringe battlegrounds

A Little Info on Early Voting
I found a great website that is attempting to track early voting turnout in the general election in detail. The struggle is that a lot of states and counties don't report early voting turnout in a timely fashion or even at all in some cases. Some states provide party and race data, some do not. But the data we do have provide some interesting insight. The complete site can be viewed at the following link:

http://elections.gmu.edu/early_vote_2008.html

Key highlights from battleground states:
Colorado -- 0.7M votes cast thus far vs. 1.0M in all of '04 -- +1% Democrats
Florida -- 2.1M votes cast thus far vs. 2.8M in all of '04 -- +5% Democrats
Georgia -- 1.1M votes cast thus far vs. 0.7M in all of '04 -- no party registry -- 35% of early voters African-American
Indiana -- insufficient data (only Marion county reporting)
Nevada -- 0.3M votes cast thus far vs. 0.4M in all of '04
North Carolina -- 1.2M votes cast thus far vs. 1.1M in all of '04 -- +28% Democrats, 28% of early voters African-American
Ohio -- insufficient data -- only 2 counties reporting

Nothing surprising in Colorado, Florida or Nevada -- they seem to be on track to have about as many votes early as in '04 without much evidenced advantage in who is voting.

Insufficient data in Indiana and Ohio don't give us a clear picture.

But just look at North Carolina and Georgia. North Carolina has a HUGE Democratic advantage, 17% larger than 4 years ago and with higher early turnout. Georgia doesn't have a party registry, but African-American turnout has been driving huge overall turnout. If I were McCain, I would be somewhat worried about Georgia and really worried about North Carolina.

What About Those Pesky Undecideds?
Bill Greener had an article today on salon.com contending that recent evidence would indicate that in a black vs. white match-up, the recent history has shown that the undecideds break late for the white candidate. He offers 4 examples from recent history where it appears all or most of the undecideds broke for the white candidate. Sort of a recent twist on the Bradley Effect. I could dispute some of his facts, such as cherry-picking the one poll in the Tennessee Senate race that Harold Ford lost that support his point vs. other polls that showed a wider spread going in. However, in some states, there seemed to be at least some validity to this trend in the primaries (notably Pennsylvania and Ohio.) So, let's at least not dismiss this possibility out of hand.

I'm pegging the independent draw at around 2% in this race, meaning that 49% will be the amount needed to win in a typical state. This is a little bit arbitrary since most of the polls don't break out "other" and "undecided", but I do the best with what I've got. Obama is currently above 49% in Nevada, Ohio, Colorado, Virginia and New Hampshire but not in Florida, Indiana, North Carolina and Missouri, where I currently show him leading.

So, in this scenario, Obama would still win if the election were held today, but it would certainly make the race a lot closer, by about 3% to 4% depending on the state.

Of course, there are also questions about the likely voter models, which some believe underestimate Obama's support by 4%, so it all might fall out in the wash.

But, so we know the range of possibilities, I will publish on November 3rd not only a final "official" projection, but two what-if scenarios -- a scenario that shows McCain getting all the undecideds and one that shows Obama doing the same.

Ted Stevens Convicted
Alaska Senator Ted Stevens has been convicted on all 7 charges. He's going to jail and not back to the senate. We can probably move one toss-up senate set over to the Dems.

ATF Foils Obama Assassination Plot
Two skinheads were arrested by the ATF for plotting to rob a gun store, shoot up a North Carolina High School in an African-American neighborhood and then assassinate Barack Obama. The two idiots probably wouldn't have survived past step one, but it is a scary reminder of just how many crazies are out there that we will have to worry about until election day and for 4 years after if Obama wins.

Trail Dust
McCain and Obama are BOTH criss-crossing Ohio and Pennsylvania today. McCain moves on to North Carolina tomorrow while Obama continues in Pennsyvlania and then on to Virginia.
Palin is in Virginia, Biden in North Carolina and Florida. Biden stays in Florida tomorrow.

Anyone see a pattern here? OH, PA, NC, VA & FL -- the five biggest prizes that are within 15 points. McCain needs 4 out of 5 to win, really 5 out of 5 if Obama carries Colorado, New Mexico, Iowa.

Stay tuned tomorrow to see if McCain keeps closing the gap....