Mason Dixon is out with a new poll, and it indicates a much tighter race in Virginia than other polls have indicated recently.
Democrat Barack Obama has taken a very narrow lead over Republican John McCain in Virginia. Statewide, 47% of likely voters currently support Obama, while 45% back McCain and 8% remain undecided. McCain held a 48%-45% advantage two weeks ago.
A poll indicating a tight race, within the margin of error, is no reason to be bummed. To the contrary, it's a bracing reality check that this thing ain't over 'til it's over. It's also all the more incentive for everyone to work their butts off the next 13 days, take nothing for granted, "leave it all on the field," and make sure that McCain-Palin don't sneak away with a narrow victory in Virginia on November 4. Thanks, and G'Obama!
UPDATE: CNN has it at a 10-point lead for Obama (54%-44%) in Virginia. "Obama's lead in Northern Virginia is growing...He now beats McCain 2-1 in the Washington suburbs."
It is unreal, however, for Barack Obama. This is the only poll released in the last few days to show Obama below 50%. So excuse me if I have my doubts. I'd like to see the poll's model.
Unfortunately, it does not tell you the weighting in terms of party but it's possible to figure that out by the numbers.
There are many things for us to be optimistic about in light of this poll. One, M-D's results have been skewing in McCain's favor. EVERY other recent poll released on Virginia shows Obama over 50 percent. With this in mind, I think it's a good sign that now even M-D is showing Obama up. Two, the margin in Northern Virginia is exactly what we need to carry the state. If Obama wins that area by a 2-1 margin, there is no way McCain wins this state. Fairfax tends to be the bellwether of NOVA. It typically falls somewhere between the outer suburbs of Prince William and Loudon and the inner suburbs of Arlington and Alexandria. A 2-1 margin in NOVA overall indicates to me that he is leading by a similar margin in Fairfax. That type of margin in Fairfax combined with what is surely a much higher margin in the inner-beltway cities would give Obama a huge lead that a downstate victory for McCain would not be enough (a downstate victory is not a given either considering the Tidewater numbers). This is especially true given the fact that even his base areas downstate is not where they should be for a victory. Obama is at the 40 percent mark in Southwest Va. and is even close in the Lynchburg/Southside area.
That brings me to my next point, relating to the 5th District race. If the McCain-Obama race is that close in the Lynchburg/Southside area, then Perriello has a good chance to pull an upset. This likely means that Republicans like Goode are not as far ahead down there as usual and won't have anywhere near the type of coattails they had going for him in '04. Also, Fairfax and Albemarle tend to deliver similar percentage margins in elections. They are typically only a point or two away from each other. If we assume that Fairfax is the bellwether in NOVA again and is giving Democrats over 60 percent of the vote, then it would probably be safe to assume that Albemarle is doing the same. Getting around 60 percent of the vote in Albemarle (which is interesting enough about 1/7th of the 5th's population like Fairfax is 1/7th of the state's) would be great news for Perriello when you combine it with the lagging Republican numbers in Southside, the 75-80 percent share he will get in Charlottesville, and the large turnout among students in C'ville and blacks in Southside. That would make the district look somewhat like '05 when Kaine managed to carry it. Kaine won with 61 percent in Albemarle, 79 percent in Charlottesville, while keeping the rest of the district relatively close.
Second, we should be sure to recognize that it does not appear that CNN is the outlier here. It is relatively in line with other recent polls.
Third, I applied the numbers from the internals which have Obama winning 82 percent of the Democratic vote, 49 percent of the Independent vote, and 11 percent (I believe) of the Republican vote and applied them to the CNN '04 party weighting to see what I would come up with. The answer? About 45.7 percent for Obama or about a point off from his position in this poll. So I added a point to the Democratic party percentage from '04 and subtracted one from the Republican total. After this, I got Obama at 46.44 percent which is similar to what M-D has him at. Based on this, it seems that M-D assumes that the party ID break down is something like 38 percent Republican 36 Democratic and 26 percent Independent. Does anyone believe Republicans still have an edge on Democrats in this state considering the 400k plus new registered voters we have? (honest question)
Forth, the poll assumes a 50-50 split among men and woman. In '04, women were 54 percent of the electorate and men were 46 percent.
Finally, the poll sample is only 18 percent black. In '04, blacks made up 21 percent of the electorate. That number is going to be at least a point or two higher this year. So, more realistically the black share of the electorate will be about 4 or 5 points higher than what M-D assumes.
I think when you consider all of these factors you can understand why they show the race tighter than everyone else.
McCain cannot afford to lose any. So, if on election day, McCain wins almost all of these by 1 point, but loses lets say Florida, or Ohio, or Missouri, he is finished. I will tell you right now, at least three of the states I mentioned will be won by Obama. That means the magic Pennsylvania, which went to Kerry and Gore have to go to McCain. McCain is down 12 points in Pennsylvania.
We can be confident, however, if Obama wins Virginia, he wins the White House, just like when Jim Webb won Virginia, the Democrats won back the Senate.
Still, now is not the time for complacency.
For my money, Pew is about the best and not just on politics. They do important work on religious and cultural trends and because they are non-partisan, they don't have a dog in the race.