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

Election Night scenario-tracking tool

October 31st, 2012, 10:25pm by Sam Wang

(Originally posted to temporary site)

Greetings, all. I’ve been a bit slow with the analysis. Disruptions in Princeton…though we do have power! It is interesting to see Governor Christie and President Obama get serious – together – about the work of recovering from what I hear is a major mess.

Reader Jeff Cardille has cooked up a cute Excel spreadsheet to allow you, on Election Night, to automatically tabulate the victory scenarios for Obama and Romney. Enter the winner as he is declared in each of 8 swing states (256 possible scenarios). The spreadsheet shows what fraction of the remaining possibilities lead to victory for each candidate. It’s neat – check it out EV-Tracker-2012 (and two Web versions) for Election Night. Soon I’ll also have a Geek’s Guide (as I did in 2008).

Thanks to Jeff C., d c, David R., Sanjeev K., and Brian C.

Tags: 2012 Election · President

13 Comments so far ↓

  • Tapen Sinha

    Thanks Sam. I will be ready with my bowl of popcorn. Oh wait, I have to teach Tuesday night….


  • Dgarr

    This is the coolest tool. Thanks so much!

  • Matt L

    Much respect to Jeff Cardille, and the two other web-devs. However, the NYT has put together a very user friendly and visually appealing version that PEC readers may also enjoy:

  • Bart

    Glad to see your site back up. You mentioned on facebook that you’d be making some adjustments to your model for making the final prediction. Can you briefly say what these adjustments are?

  • Marvin8

    Pardon my ignorance, but I thought there were only 538 electoral votes. Several of the scenarios on the election night tracker total 553 votes. What am I missing here?

    • Sam Wang

      Oh, dear. Which tracker and which scenario? Can be fixed. Please help!

    • Dale

      I thought I also posted a comment about that problem, but it didn’t show up.

      It’s the “2012 Election-Night Helper”. And the problem is with the pre-count. In other words, the starting totals assumed for the solid states for each candidate is too high to start with. As the swing state numbers are added in, the final number is always 553, which, of course, is too high.

    • Sam Wang

      Report it to Brian. His email address is there. I think he’s starting from the original EV count in Jeff’s Excel spreadsheet. Therefore he is counting NC in the baseline and in the tossup total.

  • skmind

    There are two things I’ll be watching, assuming things hold as they appear to be, i.e., Obama winning on Tuesday

    1. How does Dr. Wang’s model perform this time, because if he gets it as close or closer than Nate Silver, again, then it’d be a good example of Occam’s Razor. Why make more assumptions than are necessary?

    2. How will the pundits spin it? I have a good idea (, so I would test my gut feeling since I am not so good at gut feels about the election (I was predicting like Joe Scarborough did, based on a hunch), and thought Romney would win 286. But that was back in June/July.

    The imp in me wants to see Romney come close and lose, though :-)

    Thank you once again, Dr. Wang. This has been an education

  • Ms. Jay Sheckley

    Will the Java maps still work?
    I’d like to play with 3 or four at once.

    • Ms. Jay Sheckley

      Ruh-rh, I have an idea that may not be possible. I’d like Java map toys that include pale blue and pink colors to keep note of who is winning where without adding em in.
      As it is with interactives you can only chose final result or white

    • Ms. Jay Sheckley

      That was ruh-roh. Will 8 am never come??? Is this the penultimate day?

Leave a Comment