Princeton Election Consortium

A first draft of electoral history. Since 2004

PEC on the BBC (and other venues)

November 7th, 2012, 11:17am by Sam Wang


At noon (in about 45 minutes), I’ll be on BBC Radio 4. At some point soon I’ll be on ‘More Or Less,’ a programme on the use of numbers in the news.

Adam Gopnik in the New Yorker. Also Mashable, Gawker, the Chronicle of Higher Education, Simply Statistics, Business Insider, NYTimes, Forbes. It’s fun to see the regular press get so interested in good statistical analysis.

Photo: Ian Holmes. Here’s a cake video. Happy birthday, Jay!

Tags: 2012 Election

266 Comments so far ↓

  • Steven J. Wangsness

    Thanks to you I nailed the outcome in my predictions, and now all my friends think I’m some kind of genius. Mum’s the word!

    • Jay

      My friend Larry referenced the PEC website, and it brought not only rational insight, it brought peace of mind on election night!!

      I could ignore the talking, yammering heads on the networks — saying stuff and nonsense just to say stuff. (Well, there WAS a lot of air time to fill.)

      I could coolly reassure my wife and others, as early returns came in, that everything was right on track, because IT WAS! No need for nail-biting or anxiety!

      THANKS, Prof. Sam Wang! GREAT job!

    • victory furniture

      Mazel Tov Sam and Andrew! It was a pleasure coming to your site daily this Fall. I will soon be suffering from PEC Presidential Election Update withdrawal but I have bookmarked your site and look forward to see what else you guys will do down the road!

  • Jon V

    I know you went with FL going to Romney, but you also said it would probably be within 50k. It looks like you nailed it. 46k difference… you were right on this election as well. Congratulations!

  • Olav Grinde

    Congrats, Sam!
    I hope you can suggest some good election follow-up stories to the BBC. Such as quantifying the gerrymandering issue…

  • SoleburyJim

    Thanks again Sam for keeping me sane with your cool logic and steady hand.

  • Matt

    Congrats to Dr. Wang, Nate Silver and all the other “liberal prognosticators” who used “math” to be damn near exactly right.

  • Khan

    Well done Dr. Wang!

  • Eric

    My calculations showed me Obama winning Florida by about 0.8 percentage points. At the moment it stands 0.6 but we might not know the final result for up to a week. In the end though I believe my prediction will hold. Obama will have won the election with 332 electoral votes.

  • Shawn Huckaby

    I’m printing my
    Wang/Silver 2016 bumper stickers this afternoon.

    • Olav Grinde

      Well, I may prefer to go with Wang/Ferguson, a proven winner. :)

    • Craig

      But if they’re both from New Jersey, then NJ electors can only vote for one of them!

    • Ms. Jay Sheckley

      Great! I’ll replace my
      OBAMA 2012 / ROMNEY 1040

      Too bad about the typo, though-
      It should read

      WANG / FERGUSON 101001e9
      They’ve Got Your Number

    • DJG

      If Professor Wang does run, it will be a first:

      The first Presidential candidate to declare victory or concede before election day!

  • Ross C

    Excellent work, good sir! Congratulations on the triumph of numbers.

    While some Republicans are already preparing for the coming conflict within their party regarding how they can appeal to minorities, youth and women, others are responding in typical fashion: we lost because we weren’t conservative ENOUGH.

    In my morning of schadenfreude, I did something I have only done once in my life: visited Redstate.com. In an article entitled, “To the Extent Conservatism Was on the Ballot, It Was Not Defeated,” we get this quote:

    “For all the talk of the need to moderate in order to win, Obama ran the most divisive, radical, and negative campaign, while Romney ran a relentlessly positive campaign with incessant promises to work with the other side. ”

    Wow. Bizarro world indeed.

    • ACBaxter

      Are you kidding me! You cannot buy, steal, and lie your way to the presidency. Obama defended himself and had an outstanding organized compaign.

    • grandpa john

      pretty much says it all about Republicans and the real world does’t it.

    • Suja

      Parallel universe. On some other planet, far, far away from earth.

    • Suja

      OY! I should just not comment when I haven’t had any sleep.

  • deckbose

    Congratulations, Professor Wang. You laid your personal reputation on the line, based on your belief and understanding in the science of numbers, and were criticized heartily for it by insular GOP naysayers. But you always kept a good humor, and I was genuinely inspired by the obvious depth of your convictions. Great thanks to you and Nate and the strength of your respective knowledge.

  • bks

    So much for the University of Colorado model.

    –bks

    • mediaglyphic

      BKS,
      i wonder if Nate Silver will also take out econometric variables. As they caused his model to be more measured than necessary. I would like to see Nate Silver do an analysis of which parameters in his model helped and which ones hurt his forecast. not sure if a simple chi squared test will do it (man its ben a while since i looked at stats).

    • Craig

      I think his econometric variables zeroed out in October. His late issue was overestimating the likelihood of systemic error by including historical elections with a low volume of polling. Naturally, the average error is going to be higher when N is small.

    • Joel

      @ Craig:

      Did Silver’s econometric variables “zero out?” I don’t really think so. He was great on the Presidency but whiffed on MT and ND senate races, largely on the grounds that he incorporated “state fundamentals” in those races. Now in ND, that made sense because there was very little good polling data. However, MT was pretty well polled and Tester was holding strong in the end. Even so, Silver went with him as a 3-7 underdog.

    • Craig

      Definitely not in the states where polling was sparse, you are correct. And of course, it should have never have been anything but zero. He’ll fix that for the next go-round, I think.

    • Grateful

      And they came up with their numbers BEFORE pot was legal….

  • Elizabeth Duvert

    Looks like the BBC is finally waking up. Its coverage of this election was really poor. Mr. Mardell just couldn’t get “the soul of a nation,” to use his own words. Will look forward to hearing you on BBC.

    • acae

      Agree: the BBC bought into the whole “horserace” story … too close to call, etc. Lazy journalism methinks.

    • Stephen Coltrane

      Mark Mardell has been terrible, far too quick to spout lazy clichés. He was the only BBC journalist who never acknowledged the polls were moving back in Obama’s favour. By the looks of him, he spends more of his time in the States sampling the local cuisine than researching his stories…

      Afraid I only caught the end of Sam on Radio 4, but I might be able to catch up online.

    • Grateful

      Allow me to take this time to share my personal opinion that I have problems with Katty Kay’s style of “journalism.”

      That is all; carry on.

    • Stuff

      @Stephen Coltrane
      Could we call someone a blowhard without also commenting on their weight? What would that have to do with his ability to report on a country?

      I’m sure most of us liberals dislike weight discrimination for the same reasons we dislike race, sexual orientation or other kinds of discrimination. Letting weight be an exception (usually to justify mocking Chris Christie) just invites legitimate rebuttals of hypocrisy.

      For what it’s worth I agree his reporting is shallow more often than not.

    • Sam Wang

      I agree. Sorry, too hard to watch everything on the threads. If there’s too much of this, as traffic declines we may have to go back into heavy moderation…

    • Stephen Coltrane

      OK, fair enough, my snarky comment was just as lazy as Mardell’s commentary. I promise not to repeat it. It’s just so disappointing that an apparently intelligent journalist (who was good when he was a Europe correspondent) can’t use his objective advantage as an outsider to give a more accurate, less vested-interest perspective on the race. I guess I’d still rather have him than RCP…

  • securecare

    “…I will not have to eat a bug…”

    Maybe we can get Karl Rove to eat a few, FOX might even cover it.

    BWAH hah hah hah

    • mediaglyphic

      Karl Rove is a bug. he and lee attwater, were nothing more than race baiters. He is one of the darkest forces in the republican party, and the sooner he leaves the sooner the party of Lincoln can start a recovery.

    • neanderpaul

      @mediaglyphic

      Obviously you haven’t watched enough of my favorite conservative, Stephen Colbert, who regularly had Rove on his show. Rove is not a bug, he is a ham loaf…wearing glasses…and he is freaking delicious apparently.

      Which means, perhaps we can get bugs to eat him?

  • Child Progeny and Supergenius

    Something to take home….smaller polls with good methodology trump big polls with bad methodology(eg Gallup). The former will have a higher variance from poll to poll but will center on a more or less correct estimate while the latter usually has a lower variance but centered on an more incorrect estimate

  • Child Progeny and Supergenius

    Re Rasmussen’s claim that polling will have to be fundamentally different…I actually agree but he is clearly not the guy to figure out how to do that.

    Eventually when polls of 500 people yield 480 unemployed or retired whites all over 40, nobody will be able to wring blood from that stone

  • James McDonald

    First, congratulations! I’ve been extolling your virtues for many years now, but even as a long-time fan I’m really impressed with the precision and accuracy of your popular vote projection.

    Second, I’d like to propose a simple experiment to gauge various pollsters.

    Given a database of the median pollsters for each population/date, it should be easy to calculate the deviation for all the other pollsters on that same population/date. (Modulo some complications with tracking polls that average over several days.)

    One could then plot a histogram for each pollster showing their divergence (in SD units) over all the polls they conducted.

    That would provide a more detailed picture of the house effect for each pollster.

    Going even further, sharp peaks in such histograms would be evidence of cooked books, not that any pollster has ever been accused of such a thing.

    Just a suggestion, now that you have idle time and thousands of eyeballs going through withdrawal :)

  • Michael

    Missed most of the BBC 4 show, but caught the last of your comments, and liked what I heard. Is there any way of hearing the whole interview, or are there no recordings made?

  • Out Here In Pasadena, CA

    Absolutely brilliant statistical modeling on squeezing the numbers which all the other media outlets refused to do. “It’s a tie”/”It’s a nail biter”/”It’s too close to call” recited the dumkopfs at NPR and the rest. Really?
    This race wasn’t too close to call at any point; even Romney evinced final capitulation over the weekend.
    Next time, put some substance to the numbers which everyone is doing today, including your comments during excellent BBC4 appearance.
    Thank you kindly for sharing your expertise and genius.

  • Rose

    Dr. Wang,

    I, too, have been lurking for months now and loving your posts, the rational discussion of the data and the issues, as well as the snarky humor by the posters! I must add my thanks to you especially, and to everyone here, for creating an oasis of sanity amidst the yammering throngs.

    I am particularly struck by the visual impact of comparing your proportional EV map with the traditional maps of the US used for most red/blue analysis.

    Since there are red voters in blue states, and blue voters in red states (which is not reflected in these maps), I wonder if it might be possible for someone to do a map that ranged from red through purple to blue, based on the proportion of red/blue voters in each state?

    Obviously, this can’t be done before the election results are known, but it would be interesting to take the 2004, 2008, and 2012 election results and present them in varying intensities of this spectrum, similar to the intensities of red and blue that you use for probabilities of who will win a state.

    This would allow us to see the demographic shift in each state (due to Latinos, population shifts, etc) over time. It might also serve as a useful tool for refuting some of the inevitable spin in the coming weeks.

    And perhaps it could be a rational way of determining where to focus our efforts to shift the balance of power in each state, i.e., the more a purple state is trending to blue over time, the more we should emphasize our efforts to support the progressive candidates in those states, and get voters to the polls.

    I truly thank you, and everyone here, for letting me share the ride! I will miss you all.

  • DK2012

    Wanted to stop in and say Thanks and job well done. Loved reading your stuff and how it kept me sane for the last month. Much appreciated.

  • wheelers cat

    Sadly ….its not over.
    http://www.nationalreview.com/articles/332766/critique-nate-silver-s-pure-reason-jonah-goldberg

    And it will never be over until we can explain the biological basis of *Their. Own. Behavior* to the Jonah Goldbergs of the world.

    • Craig

      Oh, NRO is where critical thinking goes to die.

    • 538 Refugee

      Basically there were so many predictions someone had to be right. I like how he throws in the Colorado economic model and uses “garbage in, garbage out”. What he fails to point out is that it isn’t hard to figure out what the garbage is.

    • RDT

      In someways the important lesson people like Jonah Golberg need to learn is not so much that “math works” as that talking to an actual cross-section of the population works.

      Yes, statistical models can be off, but the reason that Nate and Sam were right, and the University of Colorado was wrong, was that underlying Sam and Nate’s models were opinions of real voters, and not just a model of the world as viewed by the modeler.

    • badni

      Very interesting. The Unskewed guy was a honestly wrong — he really thought he was doing something statistically valid, based on his “assumptions”.

      I have a lot of respect for the ability to be disillusioned when the evidence turns out that you were operating under an illusion. I bet most of his readers are taking the election results as evidence of voter fraud, not as evidence of unskewing being wrong.

    • Shawn Huckaby

      Zen Koan for the GOP:

      “How many times can the wrong lesson be learned?”

    • Pat

      Wow, love how he even recognizes that Rasmussen was wrong!

    • Ross C

      Wow, I’m doing lots of new things this election. First, I visited redstate.com for a morning cup of schadenfreude — my first, and probably last, visit. Now, I’ve clicked into NRO, also for the first time. I’ve always refused to click on links to them on principle. Now what in the world could have made me so suddenly broad minded?

    • Futurist

      Hey wheelers cat,

      It will never be over as long as some folks think that you can make water run uphill if you just pray hard enough. How hard do you have to pray? Why, hard enough to make water run uphill, of course!

      They want what they want, because they want it, and think that constitutes natural law. In the case of Jonah Goldberg, he starts out OK (assumptions do make a model – that, plus correct mathematics), then gets snarky (“86.6% you fools, not 86.4!” and deliberately obtuse about how to portray the results of calculations. But at the end he loses it – implicitly returning to “gut feel”: “I like to think…that the soul…is not so easily number-crunched.”

      Well, you can like to think so all you want, and if you pray hard enough, well gosh, maybe it will come true.

      Meanwhile back in the real world, I think we have to do what scientists and empiricists should do: Which hypotheses were best at fitting the result? It will be interesting, for instance, if Rasmussen returns to the same, bad old ways next time, and if the Colorado model is trotted out with the same assurance in four years. Sam, Nate, et al did a much better job of explaining the data. Dean Chambers understood that, and kudos to him for doing so.

      The lesson, I think, is: get the facts, learn from the facts – but don’t be so naive as to expect everyone else to do the same.

      Thanks, Sam, for all your work – and I’m really glad you didn’t need to eat any insects!

      Would love to see a disassembling of the results, plus the pollsters and methodologies, as a follow-up.

  • sjw

    For the past two months this website has been a necessary complement to my blood pressure medicine. Many thanks for the hard work and the expert analysis. (By the way, I’ve made PEC and 538 a formal part of the discussion in my history seminar next week. Paper assignment already handed out …)

  • Shawn Huckaby

    2004 popular vote margin: R+2.4% (3,012,166).
    Karl Rove called it a “clear mandate”.

    Once final results are in this time the margin will be nearly identical, and this with depressed turnout in the northeast. Perhaps Mr. Boehner just needs Rove to explain his definition of mandate to him.

  • dan coplan

    Superb job. You and Silver made this election
    enjoyable and memorable.

  • Jefflz

    I am an enthusiastic fan of both of Sam Wang and Nate Silver and their efforts to reduce entropy in the chaotic universe of pollsters. But as I watched the news and learned of the long lines of patient voters who wanted to express their electoral will, I began to wonder if the Uncertainty Principle might not apply to poll data analysis. If all the voters had the confidence in the accuracy of Nate’s predictions that many of us had, would they have turned out to vote? I abhorred the “too close to call ” fakery of the mass media, but did fear motivate reality?

    • Ms. Jay Sheckley

      Jefflz- Dr Wang/PEC is better than Silver: a smoother ride with greater certainty earlier. That’s why I migrated for this election.

      November 6th was my birthday, and with the help of PEC I didn’t go nuts beforehand. I knew my wish would come true.

      Sam, it’s not numbers, but what do you think re Jefflz question? Voter confidence vs Voter terror vs “all the ingredients are already in the cake.”

      Speaking of confidence that “all the ingredients are already in the cake”, here’s a snapshot from our Nov 6 celebration, at the silly tv moment where at right Romney is “projected winner” while at left (under RACHEL MADDOW) Obama’s win is announced.

      (Let’s all write Maddow demanding that Sam Wang be on her show!)

      Notice the PEC MEDIAN EV ESTIMATOR and EV HISTAGRAM, already on the cake? :D Twice the statistically projected fun!!

      http://www.facebook.com/photo.php?fbid=4925151691240&set=a.1154730673071.23076.1371822976&type=1&theater

      Photo by PEC convert/co-host scientist/UC Professor Ian Holmes, who loved seeing probability distribution charted on a cake. Frosting funded by PEC missionary & multi-year commenter Jack Rems/ Dark Carnival.

      Closeup of “PRINCETON VICTORY BIRTHDAY 2012” cake at https://www.facebook.com/photo.php?fbid=4923222363008&set=a.1154730673071.23076.1371822976&type=1&theater

      So how should we approach Maddow?

    • Kenn

      My very unscientific observations persuade me that people very seldom cast votes on a rational basis. They are motivated by tribal, atavistic, semiotics.

      My Inbox was filled with laughably false spams from conservative friends who claimed with a straight face that the President wanted to replace the National Anthem with “I’d Like to Teach the World to Sing,” a Photoshopped picture of the President’s wedding ring claiming it was Muslim calligraphy, and in the most ridiculous one: an Onion article claiming the President had an illegitimate son. When I debunk them, I’m met with silence, and only in the case of the Onion article did the sender reveal to his mailing list that he was mistaken.

      My point is, if these people (all angry whites, by the way) had the least inkling of integrity, they could have spent 30 seconds doing a web search to test the veracity but their blood hatred of the President blinded them to all reason. (One of my senders claimed to have been a card-carrying Mensa member!)

      Do you think the mass media monolith was disingenuous or simply biased toward the “dead heat” angle because it’s more interesting. (Contrast this to the usual accusation of liberal media bias. In this case, it was biased in favor of the conservative!)

      I also think there was a decided bias in favor of simplicity. It’s much easier to understand national poll numbers than spend the time explaining to a math-phobic audience the significance of the state polls.

      My evidence, again unscientific, is every time I meet someone who hates the Electoral College, when I talk about its origins and how because of Constitutional reasons, it would be impossible to get 2/3s of the states to approve an amendment abolishing the Electoral College, I get blank stares. It’s easy to have an opinion but much more difficult to put any thought into its formation.

    • wheelers cat

      Ms. Jay, im ravished.
      there are no words.

  • DCWoman

    Thank you Professor Wang. You were my calm this past week.

    For someone so smart, you’re actually very funny. I find that odd!

    Your mother should be proud!

  • Paul Garvie

    Sam/Andrew:

    I can’t thank you enough for the work you’ve done here. You’ve been an incredible source of information (and assurance!!) throughout.

    What a long strange trip it was!

    See you in 2014, gentlemen.

    PaulG

  • Scott Tribe

    There’s a post at Daily Kos with a chart done by Fordham University analyzing the pollsters and who was most accurate (and least). “Democratic-leaning PPP” (which is how they get described by pundits and opponents) came as the most accurate. (Actualy top 2: you’ll see why there).

    Interesting to see who is at the bottom tier:

    http://www.dailykos.com/story/2012/11/07/1158157/-Most-accurate-national-popular-vote-pollsters

    • Joel

      Kos is gloating too early. The total votes have yet to be counted. And the real measure has to be comparing state projections against state results, not just aggregating and comparing against national results. Anyone knows that’s an unnecessary loss of precision.

      That said, PPP looks like it did very well and will probably still look very good after all the votes are counted. I also agree that including cell phones is still not sufficient, because many people (like myself) aggressively screen cell phone calls.

    • 538 Refugee

      That list would never fly on this site because it lacks the relevant numbers. ;)

  • paul griner

    Glad to see so many interviews in the aftermath, but I haven’t yet seen this: anyone yet swing an interview with Ted Nugent in jail?

  • Khan

    Heitkamp won in ND!

    • wheelers cat

      And Mia Love lost in Utah. They teach us that in evo theory of culture.
      Race trumps religion.

  • A New Jersey Farmer

    Mia Love was ahead in some late polls, so it was nice to see that they were wrong.

    And speaking of wrong, how about that Rasmussen fellow?

  • Jeff

    I think this site isn’t giving credit where it’s due… http://isnatesilverawitch.com/

  • melancholy aeon

    I think it’s time we give props to a conservative person who was exactly correct about the election last January. It was in fact Ann Coulter who correctly predicted that if Romney was nominated, he would lose by a considerable margin.

  • h

    What I find most interesting is not that the projections here were right, but that were right so long ago, and nothing ever really changed. A little temporary blip around Ryan choice and a slightly larger, but also temporary blip after debate 1; but neither blip large enough to change the bottom line conclusion.

    So a serious objective news organization would have said in (I don’t know) March “President Obama’s re-election seems secure.” And then would have nothing new to say until this morning. (obama wins, just as predicted.)

    This is a reminder that news outlets don’t particularly want the truth, they want viewers (readers); the truth can be boring, so to keep readership/viewership they need to start making things up.

    • E L

      Has “And the truth shall set you free” been replaced with “The truth would bore you to tears”?

  • ruth37

    Votamatic had the race pegged at 332-206 since July, with barely any change throughout the rest of the campaign. Basically, that was all the states Obama won in 2008 except North Carolina and Indiana. http://votamatic.org/

  • Shawn Huckaby

    BTW, I don’t know how you’re funding this site, but I’d sure be happy to click on a PayPal button now and then.

    • Rose

      Absolutely–me too!

    • Michael

      Yeah. Maybe we could fund some glitzy upgrades for the next cycle. Definitely some interactive tools. Maybe heat maps? I hear the cool kids all love those.

  • Joe from Florida

    I figured FL would go to Obama when you had it as Tie. You’re giving the coin-toss to Romney was just the usual Liberal pessimism… You should have realized that any poll bias was going Romney’s way.

  • Olav Grinde

    By the way, here is an excellent political analysis by Robert Shrum — perhaps the best I’ve read in the aftermath.

    http://www.thedailybeast.com/articles/2012/11/07/election-2012-myths-lies-and-losers.html

    • wheelers cat

      Olav, i left a comment for you that wound up in moderation. bien merci pour te compliment.

      Do you think Rove and Romney refused to concede because they were hoping for Michael McDonalds Nightmare Scenario in Ohio?

  • MAT

    A pretty decent mea culpa on the whole polling ‘oversampling’ issue:

    http://hotair.com/archives/2012/11/07/rcp-poll-averages-predicted-49-of-50-states/

    • wheelers cat

      quelle horreure MAT.
      NEVER link HotAir at PEC except to mock them.
      This is the tail end of the the twitter fight between Linzer, Trende and Jay Cost.

      Allahpundit and Trende are the lowest– they are actually intelligent enough to know better.
      They are conservative claquers for profit, and not ideology.
      /spit

    • MAT

      @cat

      Well, okay. But reading it brought a big smile to my face, as that article was the endgame to a long, long conversation I’ve been having with a lot of my conservative friends about why places and methods like PEC are much better for their overall sanity, regardless if they liked what the Meta-Margin was telling them. The towel has been thrown in.

    • wheelers cat

      riiight MAT

      this is purely disgusting.

      “wait until political junkies fully absorb the Jackman/Silver/Linzer record this time. You’ll need a statistics degree to follow elections soon.”

      oh yes, those intellectual elistist egghead snobs.

      Allahpundit is an effing panderbot. he sold himself to Michele Malkin years ago.

    • wheelers cat

      yeah.
      allahpundit.

      If you have to endure slavery, be grateful that your masters have ancient wealth;

    • E L

      @cat I find the comment about needing a statistic degree weird. I have no, zero, real knowledge of statistical theory and yet I easily followed Sam and Nate. Man are these guy dip… oh, well, this is a civil culture here.

    • wheelers cat

      E L
      “I find the comment about needing a statistic degree weird”

      its IQ baiting. like race baiting but with braaaiiiinnnss.
      they do it all the time.

  • Emory Mayne

    Congratulations again.

    I wonder if even you, Prof. Wang, could possibly quatify the importance of your work to general public, and its impact upon American society.

    Cheers,

  • Martin

    Sam, Nate and who ever runs Votomatic ought to be on the ELECTION FORECASTERS MOUNT RUSHMORE”

    Sam got the gold in 2008. If Fl goes O i’v got to give the gold this year to Votmatic then Nate gets 2nd place this time but it’s a photo finish between Nate and Sam this year.

  • piktor

    BBC has a list of 17 news outfits that mention Sam Wang:

    http://www.bbc.co.uk/search/web/?q=sam%20wang

  • wheelers cat

    I have to say something here.
    There is nothing that I destest more than the inter-pundit “professional courtesy” afforded to intellectual whores like Allahpundit and Sean Trende.
    im done with it.

    “Much have I said before to serve necessity,
    But I will take no shame now to unsay it all.”

    yak yak yak.

    • wheelers cat

      that would be, *cheap* intellectual whores.

    • Jack Rems

      @wheeler:
      ??
      Was this a reference to something that’s been deleted?

    • Muhahahahaz

      Yeah… somehow that comment keeps moving to the bottom of the page, even though it was posted a few hours ago.

    • Ms. Jay Sheckley

      Good to know, Muhahaha-e9
      Yet Jack seems to have escaped this fate.

    • Shawn Huckaby

      I believe it was directed toward a certain Hedgehog…

    • Ms. Jay Sheckley

      Shawn Huckaby wrote about wheelers statement,”I believe it was directed toward a certain Hedgehog…”
      I hit the reply button right under Shawn’s comment. Let’s see where my comment
      [THANK YOU, SHAWN]
      ends up.
      PS wheelers cat, if you call Hedgehog an intellectual whore, isnt that unfair to working grrls? Methinks him more an intellectual fraud [We’d call him a c— but he lacks both the depth and the charm]

    • Tim

      Awesome cake video.

      What bookstore did you mention right at the beginning?

    • Shawn Huckaby

      On that topic, perhaps a housekeeping suggestion going forward: would it be possible to adapt the comment section formatting to allow multiple nested replies (ie a reply button on every post)?

      I too have had problems with misplaced rejoinders!

    • 538 Refugee

      I’ve suggested, and received favorable response, to trying out forum software. We’ll have to see how that works out. With forum software you just quote and the reply with the quoted message end up at the bottom. Usually they have a feature that remembers where you left off so you can start at that point. I don’t mind going through a few hundred posts. I just don’t like going through them 10 times or more looking at time stamps, especially when they start getting split up on different pages.

    • Steve_OH

      I like the way that DailyKos does it, where new posts (where “new” is defined as “since the last time you loaded the page”) are marked with a big red NEW so that you can easily find them buried in the thread.

    • Jun Talabucon

      I do know that the election was over… but 2 weeks before the election, the hype was all about the (9) swing states and where these states stood in the polls, specifically, OHIO. My question is:

      What are the odds wherein Pres. Obama overperformed the pre-election polls in 8 states… and underperformed in OHIO…?

      And then, Karl Rove, even to the point of embarassment in national TV, was almost certain that the President will…

    • Elliot

      Here here

    • Hedgehog

      Ha, I’m flattered I’m so much in your thoughts. And here I thought only Wheeler was so obsessed.

  • Mary

    Thank you for keeping me sane, Prof. Wang!

  • bks

    I think we should give mention to
    http://www.colleyrankings.com/election2012/
    who got close (304-234) with a model so simple that a smart 10th grader could implement it with paper and pencil (take median of the last months’ polls for a state, add EV to candidate’s total). Just doing that would have made you smarter than Limbaugh, Barone, Krauthammer, Morris, Rove, Will, etc.

    –bks

  • George Collier

    Sam and team.

    You helped me keep calm and carry on.

    Thanks so much!

    Science rocks and you just proved it! Looking forward to your post race analysis of the various models.

  • 538 Refugee

    Been meaning to ask if tags work. Find out in a second…… ;)

    More numbers

    • 538 Refugee

      That data-driven decisionmaking played a huge role in creating a second term for the 44th President and will be one of the more closely studied elements of the 2012 cycle. It’s another sign that the role of the campaign pros in Washington who make decisions on hunches and experience is rapidly dwindling, being replaced by the work of quants and computer coders who can crack massive data sets for insight. As one official put it, the time of “guys sitting in a back room smoking cigars, saying ‘We always buy 60 Minutes’” is over. In politics, the era of big data has arrived.

  • Pat B

    Thanks again, Dr Wang and Andrew for keeping me sane through trying times. Bill Foster (IL) used to be my congressman and was really great. I am so glad he won in his new district. Congress needs more scientists. You are the best!

    It has been a treat to read all the insightful comments – even poetry and song – with this group. Hope to see you in 2014 or sooner.

  • Kathy

    Hi Dr. Wang,

    So thrilled you were right on the money! Impressive!

    Would love to see your stats on the chance of Hillary Clinton running (and winning) in 2014. And, what’s your recipe on getting kids interested in mathematics?

  • Muhahahahaz

    What do you guys think are the chances that Obama will win the PV by +3.0% or more?

    • wheelers cat

      small.
      Michael McDonald forecast the 2012 election turnout will be 2-2.5 % less than the 2008 turnout.
      mostly due to Sandy,

    • Muhahahahaz

      I extrapolated the current numbers from each state, and I came up with the following.

      Obama: 67,580,339 (51.553%)
      Romney: 63,508,751 (48.447%)

      For instance, in California the current count is 5,573,450 to 3,635,571, but that’s only estimated to be 69% of the vote.

      (Yes, some sites mention that “100%” of CA is reporting, but that just means every precinct has returned some portion of their votes. They have until December 7th to finish counting all of the votes, in CA.)

      If we assume that the current percentages will stay about the same, e.g. 59% to 39% in CA, then we can estimate how many of the remaining votes will go to each candidate. For instance, in CA this would give a final total of 8,077,464 to 5,268,983.

      Doing this for each state, we find that there are about 13 million votes remaining, with about 7 million going to Obama and about 6 million going to Romney. Of course, these assumptions may be wrong (and the “percent counted” estimates could be off for some states), but this leaves us with the totals above. :D

      Anyway… does this seem reasonable? Discuss…

    • Olav Grinde

      @Muhahahahaz: What about provisional votes — for all the states?

      Do you have a list of numbers? I haven’t seen it, and I am surprised that this very important statistic does not figure prominently amongst the numbers released.

Leave a Comment