Princeton Election Consortium

Innovations in democracy since 2004

Outcome: Biden 306 EV (D+1.2% from toss-up), Senate 50 D (D+1.0%)
Nov 3 polls: Biden 342 EV (D+5.3%), Senate 50-55 D (D+3.9%), House control D+4.6%
Moneyball states: President AZ NE-2 NV, Senate MT ME AK, Legislatures KS TX NC

McCain finishing strong?

November 3rd, 2008, 8:41pm by Sam Wang

John McCain had a good weekend. He had sharp upticks in his support in close states including Indiana, Missouri, North Carolina, and Pennsylvania. (Actually, PA went up mid-last week -Sam) However, it doesn’t appear to be enough. At the current rate, he might be caught up…by Thanksgiving. The good news for him is that assuming these last polls are predictive, he may well avoid the ignominy of a 2-1 defeat in the Electoral College.

At this point we have ties in North Dakota, Indiana, and Missouri, and a neck-and-neck race in North Carolina (McCain +1%). These are enough to shave off some of Obama’s lead. The current bottom line? Obama 352 EV, McCain 186 EV. The Popular Meta-Margin is Obama +6.3%, a distinct downtick but still within the range seen for most of October.

The current probability map looks like this:
Today's electoral college map

As of this writing, the safe EV counts (snapshot win probability >95%) are Obama 311 EV, McCain 157 EV.

If we add 2% across the board for McCain, the map looks like this: Today's electoral college map

Still no joy in Mudville for McCain-Palin.

Full analysis, including the bias calculation, in a few hours. It’s time for sushi!

Tags: 2008 Election

13 Comments so far ↓

  • Tapen Sinha

    Don’t know if you have seen the latest
    on Bradley Effect

    It would be nice to see a 3-4% added
    rather than 2%.



  • Amy

    I have thoroughly enjoyed this website for the last couple of months. A physician associate of mine, a Princeton grad, told me yours was the best site for accurate analysis. I hope the analysis holds for another 24 hours!

    thanks for advancing this old white woman’s education! Amy, Vienna VA

  • Kaz

    Dr. Wang,

    So what is your read of PA? Do you think that the last few stumps around in that state by the McCain campaign has gained them sufficient ground to give their campaign cause to believe they may win there?

  • Keith

    Dr. Wang,
    Thanks so much for putting this site together. I’ve enjoyed it for 2 presidential elections now… But I’ve enjoyed it much more this time as it’s quite a bit more reassuring.

    I just got back from canvassing and I gave money to your actblue page a few weeks ago. Thanks for inspiring us all to do our part and providing reassurance to boot.

    Boulder, CO

  • Joel


    Go to the interactive map. Click on the “with 2% for McCain” link.

    You’ll see that, even granting a 2% advantage to McCain, PA is still blue.

    You can do your own research here, Kaz.

  • dan

    Mac is back!

    Or it is the tightening that usually happens.

  • Nicholas J. Alcock

    Dear Sam,
    In 2004 you allotted “Undecideds” largely to Kerry. Interestingly recently , a pollster said most “Undecideds” are white, male, less educated. Strangely. these voters are traditionally, Republican. So, whatever model pollsters use to allocate “undecideds” they should look at their background, their culture.
    In 2004 Karl Rove identified likely Rep voters by their gun ownership etc. This was very clever and perhaps more sophisticated than pollsters as cultural identification is more important than party ID.

    Now, I suspect the game is done. But, there are many issues for SW, AM, MB, KN , AB et al to contemplate???

    May I suggest for 2012 that early voting statistics benchmark
    1)ev’s as a % of RV.
    2)ev’s as % of LV voters
    3)ev’s as % of 04 voters
    4)ev’s as % of 08 voters

    Maybe Mason Univ and EVIC atReed Coll, ORE.
    will have benchmarks to judge early voting turnout and its significance?

  • IYH

    Sam, please add me to the chorus of thank yous. This site is, obviously, a tremendous service and one that is very much appreciated.

  • Vicki Vance

    Sam! Please just reassure me that it will all be OK. And, yes, I know that you are trying to be straightforward. And, yes, I know that I am being irrational. But…I don’t like when you call McCain “Mac” – it seems too friendly, too positive, too pro-McCain. AND I don’t think I can go on if Obama loses. And…I am just too tired and too stressed. Thank God it will be over soon.

  • Sam Wang

    Vicki Vance – That’s what his supporters call him. Hey, why not take a break. There’s not much to see until late tomorrow afternoon. We’ll still be here.

    IYH and everyone – thanks for making it fun.

  • Ed

    Sam, what should we see late tomorrow afternoon? I thought we would have to wait until polls close on the East coast to get the first news from FL, NC, and VA. Will there be an earlier source? Thanks!

  • Sam Wang

    Ed – I’d watch NH, the logic being that it’s a small, quickly reporting state. The current polling snapshot there is Obama +11.0 +/- 1.5% (median+/-SE). Compare this with actual returns to see if there’s any bias. If Obama is leading by 8% or more then he is highly likely to win.

    I’ll post some poll closing times. Here are some early ones (Eastern time):

    6pm: IN, KY
    7pm: VA, FL, GA, NH, SC, VT
    7:30pm: OH, WV, NC

    For monitoring purposes, IN (tie) and VA (Obama +4%) should be good too. VA’s a bit tricky because it’s quite heterogeneous and early returns may be misleading.

  • Jeff Stake

    Yes, a map with a 4% boost for McCain would be helpful.
    Many Thanks for all you’ve done!

Leave a Comment