Princeton Election Consortium

A first draft of electoral history. Since 2004

Presidential prediction 2012 (Election Eve)

November 6th, 2012, 1:47am by Sam Wang


If state polls perform as well as they did in 2004 and 2008, most aggregators should get within +/-15 electoral votes and 48/51 races correct. The question is how to squeeze a bit more out of the data.

The topline listed just below the title of this website is not our prediction, but the automatically-generated snapshot. It will fluctuate as the last few polls trickle in. The last 3 updates are at 8:00am, 10:00am, and noon. Then we freeze it.

I make two electoral predictions.

ELECTORAL PREDICTION (mode): Barack Obama 332 EV, Mitt Romney 206 EV. The mode is the single most frequent value on the histogram. It corresponds to the map below, and has a chance of being exactly correct.

ELECTORAL PREDICTION (median): Obama 309 EV, Romney 229 EV, Popular Vote Meta-Margin Obama +2.34%. This is  the automatically-generated snapshot for November 5th 8:00pm. This prediction is almost guaranteed to be off, since 309 EV is not a common combination. (Note: I’ll edit this with the last few updates.)

ALL-STATE PREDICTION (binary outcomes):

Most outcomes arose clearly from the median of the last week of polling. The exception was Florida, for which I used polls with at least half of their respondents on Nov. 1 or after. This gave a median margin of Obama +0.5 +/- 0.8% (n=10), for a win probability of 72%. I will be unsurprised for it to go the other way (outcome Obama 303 EV, Romney 235 EV). In Florida, a recount is triggered by a margin of 0.5% or less (recount rules, Brennan Center for Justice). There is about a 50-50 chance that we will see that happen.

The next-closest states are Colorado, Virginia, and North Carolina, with margins of less than 2.0%. Excluding these three states and Florida still leaves Obama 281 EV, Romney 191 EV.

In addition to all-or-none outcomes, later today I will provide final polling medians in graphical form in the Geek’s Guide. I hope it will be useful as you follow the returns.

POPULAR VOTE. The average Meta-Margin for the last three days was Obama +2.7%. The median of national polls is Obama +1.0 +/- 0.5% (n=13 polls). The approach I described before for combining these measures gives

  • Final predicted popular-vote margin: Obama +2.2 +/- 1.0%.
  • Two-candidate vote share: Obama 51.1%, Romney 48.9%.
  • Allowing 1% for minor-party candidates: Obama 50.6%, Romney 48.4%.

Finally, here is a look at the histogram as it stands tonight. The EV histogram has resolved to just a few peaks because so few states are in play – the four states I mentioned before, plus maybe Iowa. I’ll update this with a cleaner version with longer-term-integrated probabilities in the morning.

 

I should point out that A note: Election Eve prediction is not the most impressive of feats. What we did in August (here, and here) was the interesting part. Those were true predictions, and were centered around Obama 315 EV, Romney 223 EV.

Put your predictions in comments.

Tags: 2012 Election · President

294 Comments so far ↓

  • Joseph

    I think the hot air is going out of the Republican propaganda machine at an amazing rate, and that the result is a statistical move towards a solid President Obama win. I see it Nate Silver’s latest posting, and I see it on this site as well. Under the circumstances, it becomes conceivable that the President could sweep all the battleground states, up to and including North Carolina. So I’m going to say 347-191, for the President.

  • Iseeurfuture

    My Two Cents:

    352: O
    186: R

    • ElegantFowl

      O: 285
      R: 206

      Ohio and Florida nullified/ignored/derided due to corrupted processes.

  • E A Dawsoni

    I’ve been considering what should pop up Nov 7th when I click on PEC’s Chrome tool, as a placeholder until we start to see the next cycle’s poling start to roll in.

    An actuarial projection of the current likelihood of a SC appointment to each seat, resulting in an end-of-term projection of likely court composition, might be interesting, if a bit macabre.

  • xian

    I think NC is in play, but that’s my pundit-gut talkin, not the numbers.

    I think there is a majority of people who want to vote for Obama in Florida but I am not confident they will all be able to vote or their intent will be captured.

    So, I’ll go with 303 undisputed electoral votes for Obama.

    I give Obama 50.8% of the popular vote, 3rd party candidates 0.8%, and Romney 48.4%

    • David

      My hope is unprecedented minority turnout. I think black voters will increase this time. By how much and enough in North Carolina, I don’t know. And it’s just pundit feeling.

    • Ralph Reinhold

      An apocryphal piece of data about the South. One of my daughter’s friend said that most of her coworkers said they are going to write in Ron Paul. The question, that I know of, was never asked in surveys. I know some, like yougov gave the option for ‘other’.

  • Bill N

    Sam,

    Do you have any confidence interval for Ohio’s +3 for Obama? I will also be sleepless in Tennessee tonight until this thing is called. I am not going to make a specific prediction but will say I think it will be closer than Sam is calling it.

    Thanks.

    • Ken

      Within the last day or so (I think) Nate Silver removed Ohio from the “Competitive” list and now labels it “Safe Obama” and closer to +4. Silver’s more conservative than Sam, so it appears that the confidence in the Ohio +3 for Obama is tight.

    • Joel

      Looks (from the map) that Obama winning Ohio is between 80 and 98%.

      Interesting that Nate is even more confident in Obama’s chances in Ohio, unless they’re both hovering at just outside of 98%.

  • David A. H.

    Enjoyed the site President Obama Goodluck and Godspeed

  • Squibnocket

    O 303/ 50.3
    R. 235/ 48.9

    • LeeBessToad

      I think Squibnocket got it about right. I have Obama a bit higher on votes, about 50.6 versus Romney 48.6, but the EV is right on. Unless Florida does come into play, then Sam’s 332 is dead on. It’s going to be one of those two numbers I think. There will be fewer surprises and less doubt than most people expect.

  • John

    Dr Wang,

    I apologize for having brought this point up before, but I wonder whether 1% for third-party candidates tells the whole story. While it may be about the right thing nationally, the ability for Gary Johnson to siphon off votes in the West may have an impact on Colorado, and Virgil Goode may well take a couple points in Virginia, which could be significant. Jill Stein, thankfully, will be a non-issue (and it remains delightfully easy and just plain delightful to shut up disaffected anti-Obama lefties by asking them how their inevitable 2000 vote for Nader worked for them) I realize it’s difficult to quantify as most polls have ignored the justifiably minor candidates, but do you have any thoughts regarding their ability to make even a modest difference in at least a couple states? Thanks!

  • Keith

    Obama 303
    Romney 235

  • Matt McIrvin

    I’m going with the 303 map, though I will be unsurprised at Obama 294 or 290.

  • kencc

    I still can’t get beyond a Jekyll and Hyde approach. As Jekyll I’m 90% happy to rationally predict 303-235 for Obama. But the contrarian Hyde in me thinks there’s far too great a consensus among the aggregators and modellers and in such cases the improbable happens. So, entirely on that arbitrary, irrational, contrarian basis, I’m 10% Hyde and wouldn’t be totally surprised if Romney gets about a 1% or so lead in the national polls which appears to give him a chance of winning with, say, 272-266 if he takes OH or similar.

  • Ari

    Obama 329. He takes NC and FL (and the other swing states), but loses OH.

  • TAW

    Popular Vote — Obama 50.9, Romney 47.9, Other 1.2

    EV — Obama 319, Romney 219

    Reasons:
    1. Obama does exactly 2% worse than 2008
    2. EV — Divide Florida Evenly. Just because.
    3. I’m not eating bugs.
    4. Green is a popular color.

  • TAW

    I don’t like Intrade @ under 70% Obama.

    People are betting money on Romney @ 2 to 1.

    Why?

    • Matt McIrvin

      There are people with money to burn trying to manipulate the prediction to pump Romney’s chances. In addition, the conventional wisdom has been heavily hedged, and at best Intrade is a conventional-wisdom meter. This has been going on all cycle. You can take their money if you’re willing to make that bet, though I understand betting on Intrade is actually kind of annoyingly hard.

    • wheelers cat

      Its Allan Marlow, Phd and Hedgehog exercising their irrational exuberance.
      ;)

    • BrianTH

      For an election-morning Intrade strategy, you could do very well just betting on all the favorites, since they tend to systematically over-estimate the implied odds of upsets.

    • Mark

      A of 9:20 a.m. – Intrade has the President at 71.5%

    • Lee

      Intrade has now moved above 70 and I expect upward drift to continue as the day moves on.

      There are a fair number of conservatives on Intrade who are really persuaded that the black swan will be sailing into view — they believe the arguments that poll internals show skewing toward democrats and that the independents are going heavily Romney.

      It’s not hard to bet on Intrade, but from the U.S. you have to do a bank wire to put money in an account. Once you’re funded, it’s easy and fun. Obama contracts on Intrade were good to me in 2008, and I’m expecting a repeat. I only play with recreational money I can afford to lose, but I’ll have to admit I’ve had to resist the temptation this year to toss a lot more money in on Obama.

  • Ari

    Reasoning for my Obama 329 (wins FL+NC, loses OH) prediction:

    Though I think there is a <1% chance of this prediction being right, nonetheless of all the precise outcomes where I won't be just one member of a big crowd making that prediction, this is the one of two possibilities with the best chance of happening. (O 300 was the other). If I pick 303, then big whoop if I get it right — plenty of other prognosticators would have picked that too. If I go with 347, it just means I think O polling is missing some voters. Not exciting, and plenty of people will make that "reach" prediction.

    But if take a stab at this outcome, with a surprise win in NC and a surprise loss in OH, and shockingly I get it right, it must mean I had the canny foresight to see the subtle signs portending a big upset in NC, and to sagely gauge the effects of voter suppression in OH. If I get it wrong, it will be lost in the mists of internet history.

    I just have to hope that this comment appears far enough below my prediction that it isn't on the screen when I'm showing off my prediction tomorrow to amaze everyone with my perspicacity. And that when Sam declares me the sole winner of the prize for best PEC site-reader's prediction, he doesn't bring up this comment.

    What's the prize this year? I hope it doesn't involve bugs.

  • Vermont David

    Prediction:

    Republicans will try to steal FL, OH, PA, CO, WI. If they succeed in all of these places then they will have stolen a third national election out of 4.

    It is harder for them to steal states with large percentages of white voters who favor the President. They can only really attack in those states where there are enough minorities and enough racism to allow for what is happening in Ohio.

    My Prediction. Intelligent, compassionate women, and above all Brave, determined minorities will save us all.

    I am ashamed as a white male to say that Romney will win white males by a large percentage. But the better angels of democracy, those groups who only really earned the right to vote in the last 100 years will take that responsibility seriously. They will stand in line for hours (paying the new “poll tax” as some have said) they will fight through racist election observers and demand that their votes be counted and their voices heard.

    America is a better place in 2012 than it was in 1912, but only because the electorate is more than just white males.

    • Sam Champion

      Wow Vermont Dave, such a dystopian vision! You really ought to get out more often. It looks as if cabin fever has set in early in beautiful Vermont.

    • Jack Wrong

      Amen, brother!

  • Osso

    Obama 343 vs Romniac 195….

    Popular vote 54% for Obama.

  • Joey Scarborough

    Chuck Todd just told me that Wisconsin might be in play despite 30 polls saying otherwise so my gut is quite upset this morning. However my gut also tells me that Rick Scott’s vote suppression efforts will have an opposite effect so…

    Obama 332 (and WI stays with Obama because I feel like Chuck Todd is trying hide something in his manscaping)

    • Mikey Armstrong

      As a Floridian I agree with this. If democrats were suffering from an enthusiasm gap in Florida, Rick Scott and his attempts at voter suppression lit a fire under their butts which is causing them to turn out in droves. Like in Wisconsin, Michigan, Ohio and Pennsylvania, the vote down here is every bit as much a vote against a Tea Party governor and his policies as it is a vote for Obama.

    • Ralph Reinhold

      WI will be one of the harder states to steal but watch lily white Waukesha County. The county clerk pulled stunts in the last four elections.

      They have open meeting laws so people can watch them count the ballots. They have loitering laws near the polls so ‘poll watchers’ can be emasculated real easy. They have representatives of both parties at almost all of the polls and counting sites.

    • Khan

      @ Mike

      Scratch MI off the list in terms of Tea Party Governor. I’m a dyed in the wool liberal and always a fan of most of Gov. Snyder’s policies. He’s a moderate.

    • Mikey Armstrong

      Khan, Snyder is no moderate.

  • AJSdownunder

    Ok this is more of a prediction interval.

    Based on the EV distribution graph: ~85% likelyhood Obama will do better than 290 (which is just more than an Ohio sized buffer – so comfortable). And about a 50% chance he’ll do better than 300. I can live with all that.

    OK apart from the game aspect, a key point abouts stats is not to get sucked in too far by specific EV numbers, but be satisfied in having a degree of prediction that is just enough to progress (or in this case get on with life without being too stressed).

  • Andy

    You know what I’ve been wondering as many people talk about a BO comeback in NC…why isn’t Indiana in play? I know…birthplace of the Klan, and all that. But BO carried it four years ago on the strength of the northwest, heavily minority part of the state, and it seems to me it’s similar to NC with its stealth population of unpolled working poor. Although it doesn’t have a research triangle. But seriously, if BO won there, shockingly, four years ago, why not now?

    • Ken

      I’m from Indiana, still visit family there, and would agree that race is a big deal there. However, it is not the birthplace of the Klan.

    • LeeBessToad

      Indiana also is very hard to know what is going on there because there is so little polling done due to state laws about “telemarketers”. I mean it’s great that if you live in Indiana you can’t be pestered with telemarketers. But in this case it sure hurts knowing what is going on. I’ve only seen 2 polls about the Senate race. Maybe it’s just what God intended.

    • Andy

      I thought it was Marion, Ind. My apologies if that’s incorrect.

    • John

      Indiana was a major center of Klan activity in the 1920s. However, the first incarnation of the Klan (Reconstruction version) was founded in Tennessee and the second incarnation (1920s anti-immigrant version) in Georgia.

  • Howie Weiner

    President Obama: 332
    Gov Romney: 206

    Like Sam I feel sure of all but Florida. But my Mom lives there and I’m counting on her to bring out the vote!

    Popular (with third party):

    Obama: 50.5
    Romney: 48.5
    Other: 1.0

    Let’s get out the vote!
    The President is here in Chicago. The party is tonight at McCormick Place.
    See you there!

    • Ralph Reinhold

      It was a mess at Grant Park four years ago, but it was great. I’m not in the city this year. I was just off the Blue Line so I didn’t care if Metra stopped running.

      It is going to be late this year, unless Obama has an early win in OH, NC, VA or FL. Then, Romney is out of the race, but they will not declare him dead until CA closes.

  • Worried Abroad

    If there is something really improbable that suggest vote tampering and flipping. Do you have the statistical means to detect it?

    I’m worried that people like this site will just throw up their hands and say “huh” and let it slide. Think of Ohio 2004 where there were tens of thousands of votes disallowed via dodgy means. Luckily, an Obama administration won’t shut down an inquiry like Bush did then, but everyone is saying a close election (numerically) which makes tampering much more effective.

  • Jefferson

    EV

    Obama: 303
    Romney: 235

    I still think Romney will somehow take FL

    Popular
    Obama: 50
    Romney:48.5
    Other:1.5

    ps – I love your site. I came onto it late (after you were on ‘Talk of the Nation’) and wish I had found it sooner.

    • Ralph Reinhold

      Friday, I would have given Romney FL. Now, I’m not too sure. Scott made a lot of people angry with the way he handled early voting and angry people vote. The ‘likely voter’ models giving it to Romney are low turnout ones. High turn out give it to Obama.

      The other factor is that Scott is unpopular. I’ve seen a 70% disapproval rating. That is why he wasn’t at the Republican convention. For some unfathomable reason, Romney asked him to tour with him today. That is going to hurt among soft Republicans and independents. It will also vote the soft Democrats.

      I’m going to give Obama 3% over the average poll.

  • BrianTH

    My prediction is the state polls exhibit a systematic bias of about 1.5% give or take (absolute value–I won’t predict the direction), which means the debate about the uncertainty that should be incorporated into polling-based prediction models remains largely unchanged for the next election (if it was under 1 or over about 2, I think the terms of the debate might change).

  • wheelers cat

    I was much less sanguine back in July.
    http://tinyurl.com/b9r6gua

    But that was before I knew about PEC.
    ;)

    • Pat

      Oh, Wheeler, are you Micah Cohen? I loooove those state by state descriptions!

    • wheelers cat

      no sillie! Im Wheelers at 538.
      In July I didnt even know about PEC.

      Micah Cohen is the 538 comment Nazi. He hates me.
      ;)

    • Joel

      @wheelers

      an aside, I hate the NYT comment section. It is one of the worst things that happened since Nate’s move. Otherwise, I congratulate him on his success (UK elections excepted).

    • wheelers cat

      Micah has deleted many of my comments and I almost got banned for talking about cell phone demographies.
      Nate may be a poll jedi (he has stood his ground amazing well lately) but Micah is a marketeer.
      /spit
      and a comment Nazi and a prude.
      He deleted my comment about HRC being too old to run.

  • Olav Grinde

    I agree with the map, with three exceptions:

    — Colorado goes red.
    — Florida and Ohio remain too close to call.

    Obama still wins decisively with 287 electoral votes, taking Virginia, New Hampshire, Nevada, Iowa…

    After the wrangling is done, final results show Obama also won in Ohio and Florida.

  • Slightly Skeptical

    Sam, could you also post confidence intervals for your official predictions? I can make them out from the snapshot, but it’d be convenient to have the official ones given (since you’ve said the snapshot and your prediction aren’t the same).

    FWIW, I’m expecting 303 as most likely.

  • Paul Griner

    Agree with Olav, though I worry that NH goes red as well. It’s a peculiar state.

    Still, that leaves Obama at 283, more than enough to win.

    And I’d be thoroughly pleased to be underestimating the President’s final total.

    Also: a note of thanks. Have been reading this site for 2+ years, and find it insightful and frequently funny, as well as a necessary antidote to the TV/Newspaper/Magazine/Radio noise. And I love the comments as well, partisan without being nasty. A delicate line, almost always well done here.

    So, many thanks.

  • mpv81

    I’m not gonna touch the popular vote prediction but my electoral prediction is:

    Obama: 290
    Romney: 248

    I hope Obama wins by more but I think this is the most realist scenario.

    Lastly, thanks to you Dr. Wang for your awesome blog on this. It’s comforted me more than you can imagine.

  • DaveM

    Obama 303, Romney 233, Paul 2.

    Just to throw in a Faithless Elector scenario as an eventual tie-breaker (and as a reminder that the prediction competition may not be won until Dec.17th).

    I agree with Elegant Fowl that we’ll wake up tomorrow with a tally of Obama 285, Romney 206, and a toxic sludge of litigation in Ohio and Florida. I think Ohio will resolve in favor of Obama, and that Florida’s controversial certification of a Romney win will be let stand by the Obama campaign in a (snicker) grand gesture of statesmanship.

  • Steve W

    My predictions:

    1. Obama wins (I’m not going to try to out-predict Sam on the specifics). Dems control the Senate but not the House.

    2. Stock market goes up

    3. Not one single rightwingnut comes on this site (0r 538) to apologize.

    4. Also not apologizing: Karl Rove, Joe Scarborough, Sean Hannity, Scott Rasmussen, Donald Trump, John Sununu, Newt Gingrich, Peggy Noonan, Lynn Cheney.

    5. I will thank my support team (cardiologist, psychotherapist, and pharmacologist) and take them off speed-dial.

    6. Life returns to normal.

    • Pat

      of course, not apologizing! Don’t you know it’s aaaaall Sandy’s fault?
      :P

    • pechmerle

      Steve, also not apologizing (for being hopelessly wrong, and a complete jerk to boot): Dick Morris.

    • Ralph Reinhold

      An interesting take on the Sandy theory. If the right buys it, SuperPACs are dead. Dumping all that money is a waste of time. Nobody’s listening to them anyhow.

  • NickT

    Obama 303 EV Romney 235
    Obama 51.5% popular vote
    Senate 55 Democrats/Independents 45 Republicans
    House 202 Democrats 233Republicans

  • wooderson

    My guess is 303 to 235 in favor of Obama

    As an engineer, I have faith in the numbers; however, I will not be comfortable until the final result has been determined. Thanks for all you do, Dr. Wang.

  • Tony Asdourian

    As of 8 AM today, looks like the mode has shifted back to 303 electoral votes, away from 332. Am I reading that correctly?

  • truthhurtsaz

    An Obama win in the popular vote and the electoral college is my first prediction. My second prediction. A certain nattering class will say it is NOT a mandate so it doesn’t really count. My third prediction. Hillary Clinton in an electoral landslide against Chris Christie.

  • garryb

    I like 332.
    But I’m really hoping for that article summarizing how the Fox pundits and other Republican gurus will explain their confidence –post the debacle.

  • Michael Worley

    I…. don’t see O taking FL and VA. And no way he takes CO… Early Voting has made that much clear

    On the other hand, Ohio is a reach.

    So… my tossups are WI and IA. Either Romney 273, Romney 263 or Romney 257.

    Potential upset: Obama takes North Carolina but not Virginia.

    • xian

      your gut told you that?

    • David

      Michael, I shall ignore the fact that you’re not basing your opinion on polling data whatsoever to say …

      I should think that if Obama wins North Carolina, he wins Virginia. North Carolina is so similar but more red. For now.

  • E L

    Thank you, thank you, Sam and Andrew. I predict with 100% certainty that I will miss you both and the community that you planted here. I first voted in Kennedy-Nixon so I hope I’m around for one more rodeo with you.

    I also know this: a pretty strong economic recovery is baked in the cake because new housing and commercial construction and consumer spending will surge markedly in the next 4 years. Obama will receive credit. The Republican party is set to explode because of internal contradictions. In 2016, therefore, I give the Democratic presidential candidate a 90% chance. The Black Swan is the dreaded and feared Bibi Netanyahu starting a Middle East war.

    And now to enjoy the great wailing and gnashing of teeth and rending of garments by the Reactionary Right in the aftermath of their can’t lose election. The Senate will be an especially bitter pill if they make no gains.

    J K Galbraith nailed the Reactionary Right decades ago: “The modern conservative is engaged in one of man’s oldest exercises in moral philosophy; that is, the search for a superior moral justification for selfishness.”

  • weichi

    Gotta love the unskewed polls guy. Railed for weeks about a big ROmney landslide, then I look at his site this morning and …. he predicts a Romney win at 275 EV???? Omney 50.67% of popular vote??? What a tool.

    • Pat

      haha yeah i just saw that! Just yesterday he was still calling Pennsylvania, Michigan, New Hampshire (even Minnesota if I recall correctly) for Romney. Now suddenly it’s all gone…

    • wheelers cat

      lol@weichi
      if he had any honor he’d commit seppuku.

      The Operative: You know, in certain older civilized cultures, when men failed as entirely as you have, they would throw themselves on their swords.

    • grandpa john

      sounds like that maybe he was just a spoofer not a real wingnut

  • Lee

    In the distribution of all possible electoral vote outcomes I see two tall peaks. I assume that is one at 323 EV for Obama. What is the other one at? Thanks –Lee.

    • Pat

      @lee
      you meant 332 i guess. The other peak is 303 (=332 – Florida).
      Actually, Sam could you clarify the reason you pick the mode at 332 when the 303 peak is slightly higher? (It’s all fine if it’s a personal prediction, but it’s contradictory if it’s meant to be based on the final histogram)

  • Bill N

    For those of you interested, I took the last ten poll results for Ohio obtained by clicking on the “Obama +3” above. I then did a random effects meta-analysis of these poll results using methods in the Lipsey & Wilson (2001) text. The mean difference between Obama and Romney was 3.1%, and a random effects 99% confidence interval for this difference is +1.3% to +4.9% for Obama. These appear to be pretty robust results, assuming I have made no errors.

  • Steve

    O 289

    Aging statistician here who has been doing (or perhaps I should say trying to do) this sort of thing since Reagan – Carter.

    Sam: You are doing a great job!

  • gregw

    Obama 51%, 294 EV
    Romney 48.1%, 244 EV
    Other .9%

  • badni

    I sure hope the poll workers in Ohio and Florida are better trained then the ones in Brooklyn. Chaos, lines down the block, ask 3 different poll workers get 3 different answers

  • george

    Obama 303
    Romney 235

  • Nathan Duke

    Voted for Obama this morning in Queens! Can’t wait for this to be over.

  • Froggy

    Obama 303
    Romney 235

    Wait, didn’t we learn anything from Jim Cramer? Let’s try this again:

    Obama 359
    Romney 179

    Tomorrow’s Drudge headline: “Arizona Shocker! Latinos flock to polls, oust Sheriff Joe, send Carmona to Senate, deliver state for Obama.” If I’m right, remember that you heard it here first. And if I’m wrong what’s the downside? You didn’t think that a frog would be deterred by possibility of having to eat a bug, did you?

    • wheelers cat

      Froggy, did you vote in the 538 survey in July when we were all using the interactive EV map thingy?
      http://tinyurl.com/b9r6gua

    • E L

      @Froggy Prediction: Tomorrow Drudge’s web page will be edged in black. If Obama wins Arizona, I’ll eat… an extra bowl of ice cream in your honor.

    • Froggy

      @wheelers cat: No, I didn’t vote in the 538 survey, but it’s interesting to look at those results now. Nobody back then thought that Obama would get even 320EV.

    • David

      Remember when Drudge blasted Hillary Clinton constantly because he was so sure that Obama would be easy to defeat?

  • Mike from Chapel Hill

    Romney will carry my state by a whisker. Sigh. Much as I try to deny the math, it is what it is. My only hope is that a higher-than-predicted Latino turnout will magically turn the tide.
    Either way, I appreciate the soothing effect the larger statistical picture–in the form of Dr. Wang’s predictions–has had on me over the past month. Thank you.
    If I may, an homage:

    Statistics. Stout oak
    In Hurricane’s whirling winds.
    Troubled? Try Wang’s grove.

  • Randy

    Obama 347, Romney 191
    O, 51%, R, 48%, others, 1%

    Obama’s ground game delivers NC by a whisker.

    Thanks so much for this site. Can’t tell you how much I appreciate your work!

  • wheelers cat

    Nate Silver has nearly 4k retweets on this.
    https://twitter.com/fivethirtyeight/status/265637956767846401
    I predict today we will see Nate Silver trending on twitter.
    The War on Pundits has begun. Nate Silver said he’d rather vote for Ebola. That is the problem with the Poll Jedi– they have no tact.

    • Marco

      Nate is actually being a bit bolder than usual. He must be really ***d off… :)
      On other fronts: Bayesian Prediction hits 100% !
      Tale that, pundits.

Leave a Comment