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

KCRW – To The Point with Warren Olney

November 5th, 2012, 4:00pm by Sam Wang

On KCRW-FM Los Angeles (and by podcast), some good discussion about national vs. state polls.

Ron Brownstein (National Journal), Sasha Issenberg (The Victory Lab), and Mark MacKinnon (Bush 2000) were very good on how polarized the electorate is. Sasha Issenberg speculated that “fewer than 10% of minds can be changed.” I’d put that number far lower. If you look at the Meta-Margin graph, the swings of this year’s race could be explained by as few as 3% of people changing their minds. That’s assuming none of the ups and downs come from changes in voter morale on either side. The swing is half what it was in 2008.

History of Popular Meta-Margin for Obama

Also, Mark MacKinnon suggested that Hurricane Sandy “stopped” Romney’s momentum. As I have pointed out here, Ro-mentum came to an end three weeks ago. However, I am starting to see the outlines of a story along the lines of “the dog ate Romney’s homework.” Maybe this dog?

Tags: 2012 Election · President

152 Comments so far ↓

  • Marco

    I need this to be over.
    I have not done a lick of work in two weeks.
    I have taken to step OVER cracks, because I don’t want to break O’s back.
    Dr. Wang: I need the random drift and bayesian probs. to be both at 100% NOW.
    I am contemplating a return to my native Italy if Mittens wins.
    I need this to be over…

    • wheelers cat

      O Marco, have courage!
      did you see my Illiad quote?
      Pallas Athena will protect us with her terrible shining eyes that see the truth.

    • NickinLA

      @Marco. Relax. This is not a 55-45 probability or even 65-35. The probability of an O win is at 98-99%.
      Make sure you vote and everyone you know vote.

  • Steve16748

    Marco, if Romney wins and you go back to Italy could you adopt my wife and I and get us across the border with benefits. We could pay a modest bribe.

  • A New Jersey Farmer

    Raise your hand if you’re thinking, in the back of your mind, “Gosh, I really hope Sam is right.”

    OK, hands down. See you all tomorrow.

  • Anbruch

    Good god, what did Zogby put in his polling water? In general, very good numbers for Dems, but then there is the strange OH Senate race with 21% apparently undecided.

  • Neus

    Gosh….This thing is putting me on a knife edge. Can anyone suggest how I can drift into “snore-zone” tonight? Fore disclosure–All I have is Jesus! And right now, Dr. Wang’s charts :)


    • Jay Bryant

      I find reading a book before I go to bed to be some help. Then I have a shot at thinking about the book rather than thinking about the election. It doesn’t always work, though. I was sufficiently irritated about the polling place situation in Florida that I didn’t sleep much last night.

    • Curtis McMinn

      It’s night likes these that I enjoy a good helping of Melatonin.

    • Prairie Pundit

      Your link takes us to an NBC Nightly News clip of Chuck Todd’s chin merkin shuffling around a bunch of brightly colored irregular shapes on a super-wide-screen TV borrowed from Karl Rove’s rumpus room.

  • Fred

    So when it the last and therefore official prediction?

  • Reason

    When does Sam give his final analysis? I see the MM dropped again.

  • Analytical

    Sam..Is MM at 8.00 PM on Nov 5, the prediction?

  • Jack Rems

    Right now on the histogram, the tallest spike is around 303 and the next tallest around 333. The spike at 333 looks fatter. Is this just my screen or does this actually mean something?

    • Jay Bryant

      It’s a result of the image rendering. The graph is 500 pixels wide but covers 180 data points (181 if both ends are included). Neither 180 nor 181 divides into 500 evenly, so some lines are two pixels wide and some are three pixels wide.

    • Jack Rems

      Jay Bryant-

      Thank you.

  • Osso


    Nate has his new numbers:

    Obama: 314.4 to Romniac: 223.6 ECVs

    Chance of Winning 91.4%

    Popular Vote: 50.9%

    • Joel

      We have convergence.

      And it looks like an even split between carrying FL or not, with NC and IN going to Romney.

  • joe

    Can’t wait for the final poll and result, thanks for this fantastic website, i wait nervously

  • Eric Walker

    On my screen, the 333 looks like 2 pixels in width versus 1 pixel for the 303; the one in the 318 or 319 range also seems to be 2 pixels wide. I don’t know where a one-EV difference could come from, so I can’t see how those could be two adjacent spikes, but maybe I’m missing something. Or maybe it’s just an artifact of the graphics . . .

    • Olav Grinde

      Whatever the width of the spike, my friend, I’m just pleased to see them well to the right of 300 EV, and nary a trace of blue on Romney’s side of the red line.

  • Fred

    Nate just flipped FL over to Obama. So now he shows 314.4 /223.6.

    That is probably not his final prediction though. It would be so sweet to get FL. I am going to put on my Peggy Noonan smug face and say that “I just have a feeling” we are gonna win FL.

  • Barb

    Have you guys seen the latest xkcd?

  • Osso


    The key development since last week is that the two consistent outlier Polls by Gallup and Rasmussen have moved to the center by five points or so in favor of the President.

  • Bruce Wayne

    My only wish is for Romney to lose with 47% of the national vote.

  • Analytical

    After reading Michael Gerson’s article in Washington Post reading, I now realize the Romney’s likely loss tomorrow is because of 1) Nate Silver, 2) Sam Wang, etc and 4) Sandy.

    I can’t believe how dumb is Michael Gerson’s neuroscience!

  • Zenger

    Osso… So you have Romney doing worse than McCain in pv.?

  • AJSdownunder

    First a huge thanks to you and all your mates for the fantastic job PEC has done in keeping everyone around the world informed on the underling picture of what is going on in the US election polls (and in sharp contrast to the shrill media).

    One suggestion for framing the final median EV so that people don’t obsess over the single number: maybe accompany it with a prediction range where we expect it to be, say the inter-quartile range. Or since we are not fussed by the upper bound – there is a 75% chance that the outcome to be above say 290 – which is a comfortable Obama margin (more than Ohio). This might be less choppy and take the focus off the median as a single number prediction.

  • E L

    Sam is right. National polls move to state polls:
    Nate Silver ‏@fivethirtyeight
    Obama gained an average of 1.5 points between 12 national polls published today. Big sample sizes. That’s a pretty big deal.

    Nate Silver ‏@fivethirtyeight
    There’s been a pretty clear shift toward Obama in national polls. Based on most recent data, he may lead by 2-3% in popular vote.

  • mk49

    Why is PA at +3.5 margin for O., when the Pollster source data shows almost a +6 difference?

  • Michael H.

    Hi Sam: Can you explain why the numbers are still changing? I would have thought that ALL polling would have been done much earlier today, and what we would now see is your final prediction. Or will that come only tomorrow? Just curious… AND am really hopeful that you previous accuracy continues, and that your model has not been surpassed by developments that you cannot control (such as the dramatic increase in cell phone usage at the expense of land lines, which could REALLY throw ALL polling off, if I understand the way it is currently done correctly. Or am I mistaken that the decline of land lines would throw off the polling?

  • Martin

    Wow!!! I never thought i’d see 538 with a rosier electoral vote total for O than Sam! Just goes to show you that in the wacky world of politics EXPECT THE UNEXPECTED!!!!! it’s been the other way around for so long now i can’t remember the last time Nate’s O numbers were higher?

Leave a Comment