Princeton Election Consortium

A first draft of electoral history. Since 2004

As the post-debate polling cycle matures…

October 10th, 2012, 9:30am by Sam Wang



Josh MarshallAndrew Sullivan, and John Podhoretz (but not the ever-sensible James Fallows) are asking: should the President’s supporters panic? Based on how campaigns generally behave, definitely not. At least wait until the weekend. I know that seems like an eternity. Let me explain.

Romney fans hope for last week’s trends to continue all the way to Election Day. The Popular Meta-Margin for President Obama hit a high on October 3-4 (+6.2%), from which it has fallen to a near-low (+1.8%) in just 6 days. At the current rate of fall, on November 6th Romney’s popular vote margin will be a 60%-40% win.

Um, obviously that is unlikely to happen. What is wrong with the reasoning?

For one thing, now is exactly the time when a shock to the campaign should start to settle out. In the case of past large swings, getting enough state polls takes 10-12 days. Here are two prime specimens, the Ryan VP nomination and the 2008 Palin VP nomination.

Ryan and Palin bounces in Meta-margin

Note that for Romney, the post-Ryan peak was the closest he’s gotten all season. It shared one reason with the current bounce – partisan enthusiasm, as I mentioned Monday.

Today’s Meta-Analysis now has fresh polls in all swing states, mainly taken 10/4-5 – immediately after the debate. That means the initial shock is well baked in. We should have the full  measure by Sunday.

The bottom (or top, depending on your point of view) should be close to where things are now. One clue is that during the Ryan and Palin bounces, the Meta-Analysis didn’t reverse direction – not a jitter – until it was actually near bottom. It did that today. Even if things move further, a minimum may be near.

Some of you have asked: Was last week’s debate was a black-swan event? Emphatically not. For one thing: the race is still well within the range I predicted on August 3rd.To quote:

On any date in Summer and Fall of 2004 and 2008, a good estimator of the November outcome was what had happened so far: a Bush-Kerry dead heat and an Obama lead. Similarly, today our best estimate of Election Day performance is the average of June and July 2012. Right now that’s approximately Obama 315 EV, Meta-margin of +3.0%….the Meta-margin is roughly distributed like a Gaussian….In 2008, the standard deviation of the Meta-margin was 2.2%. Therefore our November prediction is Obama +3.0 +/- 2.2% (1 sigma).

As of today, October 10, the Meta-Margin is at Obama +1.82%. That is only 0.5 sigma below the predicted mean. To paraphrase the supermodel Linda Evangelista, we don’t wake up for less than 1 sigma. (Update: hmmm, lots of you are clicking that. Try this.)

That is not to say last week wasn’t surprising. The fact of some change after the first debate is no surprise, at least to readers of this site. But the size – I admit, that caught me unawares. In retrospect, though, the real surprises have been (a) how high Obama was before the debate (+1.5 sigma), and (b) how stable the race has been – even now. The actual distribution of Meta-Margins since June 1 has been Obama +3.3 +/- 1.1% (mean +/- SD). In other words, even including this week, the race has been only about half as movable as Kerry v. Bush.

As I pointed out during the late Pleistocene in early August, re-election races move within a fairly narrow range of opinion. The 2004 race, the first that I covered, spanned a range of barely 100 EV:

Median EV estimator from 2004 race

Median EV estimator from 2004 race

That year featured the fairly surreal Swift Boat Veterans for Truth campaign. In 2012, Obama/Romney supporters are in a freakout over this:

History of electoral votes for Obama

In summary, to both sides I say: Get. A. Grip.

Let us borrow from Heather Parton, a.k.a. Digby:

In this clip, you are Nicholas Cage.

Based on the short-term prediction calculation that we set up on September 29th, Obama’s current re-elect probability is 93%. And yes, you know what I think about InTrade. They are as hysterical as Chris Matthews.

Where is the race headed next? I don’t know. I do know that our brilliant media can only focus on one thing at a time. That one thing is about to be Biden v. Ryan.

Tags: 2012 Election · President

71 Comments so far ↓

  • Steven J. Wangsness

    It certainly is comforting to read this. It’s hard to believe that one 90-minute event, even if it occurs in front of 70 million people, could so profoundly affect the overall course of events. Now, if I could just get my friends to rein in their panic…

    • Sidra

      With so much at stake in this election, I am trying not to panic. My concern is (Democratic) voter apathy. I know there are a lot of my fellow dems who are going to vote, but I also know there are a lot who will not. I urge friends and co-workers and even strangers who I know are Dems to get out there and vote and remind them that an uncast vote for Obama is just one more for Romney. Breathe, Sidra, breathe.

  • Ohio Voter

    Long time follower, first time poster. Outstanding and superb post and analysis here. Forwarding this to all my friends that are currently toeing the ledge.

  • Richard

    I think a lot of readers want something that Professor Wang can’t give, which is certainty. We all have to wait for the data, but so far the race remains remarkably stable and in Obama’s favor. Yes, he could lose – one in fourteen events happen. On average they happen one in fourteen times.

  • Louis in NY

    Sam,

    I’m curious why the Meta-Analysis didn’t move or reverse direction until it was at rock bottom? Is that due to the manner/timing the data is entered or processed? I assume whatever the reasoning it’s just the way the model was built.

    As to the freak-out on the part of the pundits, I reckon your background neuroscience might have something to say about human nature :) As such, here’s a piece in the LA times on the advice Quintus Cicero provided his brother Marcus Cicero on his election in 64 B.C.
    http://articles.latimes.com/2012/jan/26/opinion/la-oe-freeman-rome-20120126

    As Alphonse Karr said, “The more things change, the more they stay the same.”

    Best,
    Louis

  • Amitabh Lath

    Sam, this case (debate #1) is unique in that we know that it was the cause of the dip (not so with swift boat etc) and that it was a not drawn out in time (a delta function in time rather than a drawn out thing).

    Other than the jobs report showing unemployment dipping below 8%, there haven’t been any other shocks to the system.

    This is a beautiful opportunity to study the time response of the system (the electorate) to a one-time shock of large magnitude, with a very sensitive instrument (ie, your median estimator).

    I would be interested in seeing what sort of time constant you can fit to the data once it asymptotes out.

    • Alan Cobo-Lewis

      Amitabh,
      Maybe. But as Sam pointed out, the system whose behavior is plotted on this site isn’t just electorate. It’s electorate + polls. So to study time-constant of electorate he’d have to deconvolve out the poll lag. Might be possible, but I’m not sure how big the errors on that would be.
      alan

  • Jim in CA

    The tracking poll at RAND tends to support your advice of staying calm. Their long-term tracking of the same 3500 people has Obama down from a 100% chance of winning the popular vote to merely 96%.

    What I find interesting is that this poll has generally followed the trends of the other polls, but in a much more muted manner. I suspect that is because the panelists have been reporting their preferences since July, so they’ve already become somewhat more involved than the general population — in some sense, their reactions to various revelations about the candidates has already been baked in to reflect the point that other people are just arriving at.

    There might be a lesson there for future campaigns.

    Just my 2 cents…

    • Matt McIrvin

      RAND has a seven-day survey period built in that smooths things about, but Prof. Wang had a point that any sharp step effect would have shown up immediately (reduced to one-seventh of its true size). The fact that it didn’t suggests to me that the RAND panel may be responding more to media coverage of the debate than to the debate itself, which is what you’d expect of people who are more politically plugged-in than average.

    • Froggy

      Another way of looking at the RAND panelists is that because they are high-interest voters, they are immune to a large degree to being swayed by certain events that sway large parts of the electorate. In other words, they are harder to bamboozle. That is what I think is going on here — the RAND panelists weren’t taken in by Romney’s moderate rhetoric in the first debate, which was unaccompanied by any real shift in his positions. They knew his positions well (better than the average voter), so they weren’t wowed by the new packaging.

  • Pilgrim

    I really enjoy your websire and your analyses of the polling data.

    Kevin Drum posted yesterday that he thought Obama was dropping the few days right before the debate. In other words, the debate didn’t cause the drop.

    http://www.motherjones.com/kevin-drum/2012/10/how-much-did-wednesdays-debate-help-romney

    Can you comment on Kevin’s post? I thought the data you been posting and commenting didn’t show a drop of 2-3 points before the debate.

  • wheelers cat

    I find the comparison of the Ryan bump, the Palin bump, and the Debate 1 bump interesting. The magnitude and duration are different but the shape is roughly isomorphic…at least, that is what happens if Debate 1 bounce conforms to a rough fractal shape.
    All three bounces seem to only have an underlying structure of base enthusiasm, ie not based on an external event. It seems to me the enthusiasm is based on a perception that (now) the candidate can win, a “game-changer”. Enthusiasm would increase party id, and response I think, in that conservatives/republicans would be more likely to take a polling call and self identify.
    Looking at RAND, again I think its unlikely that humans change their minds and go from vote-Obama to vote-Romney, but certainly the small gains Romney has made this week could be attributable to increase in strength of support and liklihood of voting.

    • Amitabh Lath

      Yes, the Ryan announcement aftermath has a similar time constant. I suspect we are seeing the resolution of the measurement (a few days) rather than the actual change of intention (probably instantaneous on this scale).

      So the question I have now is: can you believe any opinion that flips on such a short timescale?

    • wheelers cat

      I think opinion can shift in RAND in intensity and liklihood of voting– two measures of support for a selected candidate. But homo sap. is unlikely to flip on the candidate himself because of the biological basis of learning.
      I think possibly the state polls will be in between RAND and Pew. Like Dr. Wang says, wait and see.

  • Ken O'Brien

    Sam,

    Next time you link to an URL that opens with a crowd whistling and cheering for a supermodel exiting a limo – or whatever that was; I clicked it off fast – warn me….I’m in my cubicle farm here…

  • JohnJacobs

    Way to throw reason on a perfectly good hysterical meltdown / wild celebration (depending on one’s political leanings).

    Guess there’s nothing to do but wait and smile :-)

  • Vicki Vance

    I’m wondering if you have any advice for the President (from the viewpoint of a neuroscientist) for how to best handle an opponent who lies about the issues.

    • Olav Grinde

      Or advice to journalists, perhaps? ;)

      I do wonder why Governor Romney’s extreme liberties with truth, and the record-breaking flip-flopping, were not the main subject of the headlines after the debate?

      Several decades ago, I do believe the media would have made a far bigger issue of that. In fact, I believe that would have been the main focus of the post-debate commentators.

      We can argue that it was President Obama’s job to arrest Romney, and he failed. That may well be — but it certainly is the job of political commentators. But somehow that has become an almost irrelevant footnote in the greater picture.

      Had the lies and liberties been pointed out more prominently, and relentlessly, the voters’ post-debate impression would have been far different.

    • JamesInCA

      Olav: Because “Romney changes position” is not nearly as good a headline as “Romney spanks Obama.” One is far more expected and predictable than the other.

      Had Obama shot back more effectively at Romney’s policy changes, that certainly would have influenced the conversation. We know this because so much of the conversation is about how surprising it is that Obama didn’t do that.

    • Olav Grinde

      JamesInCA: True, “Romney Changes Position” doesn’t cut it. However…

      “Thou Shalt Not Lie” would be a very eye-catching and accurate headline on a article that in detail lists how a candidate has taken extreme liberty with truth.

  • Dgarr

    Thank you Professor Wang for your work here and your brilliant analysis.

  • HJM

    Setting partisan hysteria aside, political professionals clearly believe an event like a debate can move sentiment enough to affect the outcome in a close race.

    The big fluctuations in the meta margin – even when they are within the CI – don’t appear to be random noise. The fluctuations are in the expected direction, given what appear to be the triggering events.

    If I understand the meta analysis, it uses multiple polls to reduce the effects of sampling error and get closer to the true EV score. But what is the substantial interpretation of the remaining error – is that random noise, or the actual margin that campaigns have to work with?

    I don’t know if you’ve answered this before, but it certainly seems interesting right now.

  • Steve in Colorado

    Can someone explain better this ‘Power of your vote’ chart? Colorado has dropped precipitously since yesterday- why would the power of my vote drop from 25 to 3.7 in only 1 day?

    • JamesInCA

      Because, today, you don’t live in Ohio or Virginia. The “most powerful” votes are those in the states most on-edge and most likely to tip a candidate over the 270-electoral-vote mark. So the more certain a state is to fall for one candidate or the other, the “less powerful” its individual voters are.

      I live in California. Like Massachusetts, Hawaii, and Texas, there is no doubt about who will receive my state’s electoral votes, so my vote has a very low “power.”

  • Brian

    Dr. Wang,

    You write, “In the case of past large swings, getting enough state polls takes 10-12 days.” It seems that, as a result, if the race were running close, and something “big” happened within a few days of the election, the lag could cause the Meta-Analysis to not accurately predict the EV total on Election Eve. Do you agree?

  • Matt McIrvin

    About 2004: A factoid that I’ve heard bandied about by worried people lately was that Kerry was further behind prior to the first debate in 2004 than Romney was before Wednesday (so since Kerry almost pulled it off, Romney ought to be well ahead now).

    As far as I can tell, this is wrong: Kerry was not nearly as far behind as Romney was. Nor, however, did he get quite as big a boost as Romney.

    I’m wondering where the factoid came from. I suspect it was from watching the Gallup tracking poll instead of state polls.

  • 538 Refugee

    The debate may be just a “sugar high” that can come crashing down. I’m in the camp that it wasn’t so much Obama’s performance as we saw a shiny new Mitt. He appeared at ease, likable and, once again, quite the moderate. His move to the center however can turn off the base that was uneasy about his new found commitment to conservatism already. His position changes are already being exposed and that could put doubt into the moderates he was hoping to win with his performance as to ‘which Mitt they gitt’.

    Obama did his job in the debate of not getting caught or drawn into some place he didn’t want to go. Romney volunteered up quite a bit of fodder on his own but put a good face on it. His move to the middle was obviously a calculated strategy to grab undecideds because he needed to do something bold. However he is the one that will look like the undecided if the Obama team and pundits do their job.

    • wheelers cat

      The enthusiasm gap just jumped to 13 points +republican. Pre-debate it had almost disappeared. Debate1 is perceived by the base as a game changer, like Palins bounce.
      If Debate1 hasnt changed the game in the longer term, then enthusiasm will subside.

    • Olav Grinde

      Indeed, it’s an interesting question whether Governor Romney might be alienating some of his base.

      Dr Wang, are the polls showing any increase swing to third-party candidates as a result of Romney’s recent overtures to moderates?

      After all, he’s even signaling that he won’t strive to do away with pro-choice laws. And that’s pretty radical moderate…!

    • Brian

      That door, I think, would swing both ways. Up till now, moderate Republicans have been watching Mitt pander to the far-right, assuming it’s all an act, ready to vote for him anyway. Now, centrist Mitt comes along to confirm their suspicion, while right-wingers tell themselves that the real Mitt will be “conservative enough” as President to warrant their vote — they’re certainly not going to vote for Obama, that’s for sure.

      Ultimately, this speaks to Dr. Wang’s central point — there just aren’t enough undecideds to sway, or partisans to convert, to overcome Mitt’s extraordinary electoral map troubles. Demographically, it was always going to be an uphill climb for the Republican in these past two races.

      Of course, there are exceptions to every rule, and it was never going to be *impossible* for Romney to win, merely *improbable*.

  • Pankaj Sharma

    Sam , Have you ever collaborated with Dr. Wes Colley from U of Alabama-Huntsville. He is an astrophysicist and Statistician. He is republican and his prediction (pre-debate) appeared in al.com which had Obama 348 and Romney 190. Off course, lot has changed since the debate. But was wondering if his methodology is different from your models.
    http://blog.al.com/breaking/2012/10/uah_astrophysicist_and_college.html

    • Brian

      There was a mention of Dr. Colley’s site and analysis on this blog a couple weeks back. Colley’s method’s greatest strength also seems to be its greatest weakness: they don’t seem to discard old poll data fast enough (they look at the past month). So, while they’re right that temporary bumps (such as the current bounce to Romney) won’t throw their prediction off, it also means that changes in “where the race stands now” take weeks to register. He’s showing Obama 348 (his model must be predicting every 2008 state plus NE-2) because his model considers Romney’s very real debate bounce to be temporary.

      Dr. Wang’s model (and Nate Silver’s too, for that matter) shows that this race has been remarkable stable around 3%. So, it would seem that Colley’s model is just precise enough to report the state of the race last month, but not precise enough to notice if last month’s picture was mid-bounce.

  • wheelers cat

    If we cant call Debate1 a black swan event, can we call it a temporary-game-changer or TGC?

  • Ohio Voter

    Gallup Tracking shows the race still a tie among likely voters, and CNN Research Poll has Obama +4 Ohio among LV and +10 among RV.

    It’s pretty safe to say if Obama can maintain a lead in Ohio or Virginia and then win either NV, IA, or CO, it’s over. Or can skip all of the above and just win Florida.

    • Patrick Draut

      I agree with taking all of the noise out and just looking at the key battleground states and the EV impact. I think sometimes too many numbers can create a loss of focus. I have always maintained that on his best day, Mitt Romney still only has a 50-50 chance, because a loss in Florida ends it all.

  • kel

    Assuming state polls are a random and representative enough sample of voters that the meta-margin is stable and accurate measure of the outcome if the election were held today, what can be inferred about the members of segment of the population whose voting preference change over time? What is going on in their brains?

    I would expect a substantial proportion of partisans on each side whose opinions remain stable. Is this idea supported by the data? If so, how do the thought processes of us “yellow dogs” differ?

  • Zach C

    I’m not panicking. I however am concerned about down ticket effects. If this is a close election then neither house is likely to swap, but if Obama wins by the same margins he was in the post-convention bump then Pelosi could very well get her old job back. Plus it could swing a few Senate seats. Divided government hasn’t worked the past two years and it won’t in the next two.

    • Jack Rems

      Actually, it might start to work, if Harry Reid dumps the filibuster and the house is maybe R+2.
      Mmm, sausage!

    • Olav Grinde

      Zach C: The outcome you describe might well be the ideal scenario. If the Democrats “control” both houses of Congress, then the GOP can continue its strategy of total obstruction, blaming all on the Democrats.

      If the Republicans have the slimmest possible majority in the House of Representatives, they would be compelled to take some responsibility for governing. I, for one, would look forward to the compromise-is-surrender Tea Partiers battling it out with their saner comrades compatriots.

  • Ms. Jay Sheckley

    We have so many polls. Be nice if they’d ask responders to name a precipitating event. Granted, the President’s performance was lackluster and tedious. Which for some of us came as no surprise! Something which remains unmentioned: The media has taken no responsibility for the enormous effect of its winner-loser pronouncements, never acknowledging that in genuine debating, lying and “presenting false evidence” are instant disqualifiers. Still no mention that the least informed people present and showmanship set the “standards” for this debate scoring. Judging by gross impressions is for god reason the antithesis of the debate tradition. Why pretend it’s moderated or reigned in at all? Why have the event hosted if the hosts abdicate all intellectual authority, while radiating charisma mistken for gravitas? I am genuinely, beyond my plain disappointment, concerned abut the specific issues of rules and the only structure being window dressing- a false sense of structure itself misinforming the public. I’m venting here because you folks might care an epsilon about the things meant. Thanks.

    • wheelers cat

      From the media POV this had to happen. Romney was circling the drain, and its too late for him to fall on his sword and have Jeb Bush or Marco Rubio come riding in a white horse.
      If it wasnt the debate it would have been something else.
      Romney came back because the MSM needed him to come back.
      Horserace uber alles.

    • Ms Jay Sheckley

      @wheelers cat If this is about keeping the narrative exciting [which cant explain zero mention of actual debate rules on lies and false evidence] if this is about that rather than Lehrer’s unconscionable ineptitude possibly resulting from enjoying the spotlight, then that bodes well on scoring today and in future alleged debates this season. It’d go a long way toward fairness if anyone here (Sam? :D) can mention these concerns in the maistream media or major news blogs. The points again: (1)Lehrer’s rules forbade questioning the other candidate. They were to simply answer the questions. In a real debate Obama would’ve scored well. More importantly (2) In real debates falsity disqualifies. Thank you wheelers at, thank you all.

  • Brian

    Dr Wang, could you comment on David Paleologos of Suffolk University(MA)? He claimed on O’Reilly’s show he refuses to poll Virginia, North Carolina and Florida further because Obama has no chance of winning there! Yet his own poll has Obama up by 2 in VA. Is this guy credible?

  • Amy Fried

    Brian, It is clearly not credible for a pollster to claim at this point that Obama can’t win those states.

    This reminds me of when pollster Elmo Roper announced in Sepember 1948 that he was going to stop polling in the presidential race because there was no way Truman could win, except that recently Obama has done better in those states than Truman was nationally when Roper stopped.

  • Ms. Jay Sheckley

    As to Mitt’s apparent swings, he is secretly as stable as Dr Wang’s prediction of a winner: Romney stated and never denied he’ll have Robert Bork select all Supreme Court appointees. Bork opposes Roe V Wade and famously calls the 1964 Civil Rights Act “an abomination”. Do look him up.

  • MinneapolisPipe

    You have to make math sexy by bringing in Linda Evangelista to help???

    Well, I say, go ahead.

  • wufwugy

    What the RAND survey tells us is that the movement of every other poll is simply “partisan response enthusiasm.” We haven’t seen a shift in opinion, but a shift in people who care to pick up the phone

  • Howie Weiner

    Last week I wrote that the entire Meta-Margin would be wiped out in one week. Well, not quite but, 75% of the margin is now gone. I agree that the full impact of the debate on the polls will probably not be known until the weekend and so it might slip a bit more. But I agree with Dr. Wang that the bottom is probably around the level it is now. And that bodes well for the President.
    The question that interests me is why there has been such a sudden drop. I can only conjecture. On the one hand, 6.18 was probably a greater margin than really existed, a result of the major mistakes (47%, etc) of the Romney campaign. That might account for perhaps 2-3% of the drop. And the rest? Perhaps there were many people who felt they wanted to vote against the President but Romney seemed a worse choice. The debate gave them a chance to say that maybe Romney would be an ok alternative.
    But I think the greater factor is that the country is looking for leadership, for someone to say that no matter how tough things are, everything will be all right. And the President seemed tired, depressed, and basically out of ideas. Romney was relentless in his attack and although continuously lying, was firm in his beliefs. And that psychology suits the country’s needs. Can the President turn it around? I sure hope so, a Romney Presidency would be a catastrophe for the country.

  • Ram

    I am in Howie camp that Obama never had 6.18 meta margin in a 47%-47% country in 2012. The fundamentals point to a 51%-49 to 52%-48% (considering it was just a 52.9%- 45.7% verdict in 2008) election depending on how campaigns are run.

    I must say Obama team has run a terrific campaign to make us all believe it was going to be a 347+ EV votes.

    My take is it is going to be a close race with more like around 51.5%-48.5% (anywhere from 270 to 303 EVs for President Obama). I would be a little surprised if Obama makes it beyond 297 EV. If Obama vins VA, he will have won OH and CO making OH is the real tipping state for the the 2012 race.

  • NY Romneyite

    Wow this comment section is an unbelievable liberal echo chamber! Facts are facts, Romney would have won an election held last Thursday through Saturday. Every poll shows this.

    Also, you can’t say there wasn’t “even a jitter” in the polls until the end of the bounce, when we can all see the obvious “jitter” on August 19th of the Ryan graph.

    And the Palin bounce ended not because the Palin bounce ended. The exact nadir on that graph has a pretty clear “LEHMAN” label attached to it… remember that? Stock market down hundreds of points many days in a row? I think that about is enough to erase momentum for an incumbent party candidate.

    Then, you go on to say that hey, this election has had really low volatility, even if you include this first-of-its-kind big move within that realm. Well yeah, but how about we look at the volatility EXCLUDING this move and then talk about it in that context.

    The meta margin is down FIVE POINTS in a week, and this after every consultant said “debates don’t usually move polls” two weeks ago. So all in, we have to start to view this as a unique sort of debate result, to have moved the meta margin 4-5x this cycle’s standard deviation!

    • Michael Worley

      Thank you, Echo.

      And if you think people in WI won’t turn in to watch their homeboy, especially after one of the most watched debates ever, you have another thought coming. If Ryan wins tomorrow, WI could flip red, OH could be a tossup and we could have a Meta Margin below 1 heading into Tuesday’s debate.

    • Michael Worley

      What am I saying? The Meta Margin could be in favor of Romney by Tuesday. All we need is .3 a day, and a good debate would be capable of doing that.

    • Sam Wang

      These points are not bad, though the flavor conveyed of 2008 is wrong. The Palin week was the only good moment for the GOP all season from a horserace standpoint. Her interviews were unequivocal bombs.

      Leave some real identifier please next time.

    • JimInCA

      On the contrary, even at the height of Romney’s bounce, virtually every poll showed that Obama would win, by varying margins.

      Since you said “every” poll, it takes just one to refute your claim:

      https://mmicdata.rand.org/alp/?page=election#election-forecast

      Obama’s odds of winning were in the 90%+ range.

    • Ms Jay Sheckley

      NY Romneyite writes: “Facts are facts, Romney would have won an election held last Thursday through Saturday. Every poll shows this.” Um, no. This site displays an aggregate of polls, the derivation is shown. Do google our host Dr Sam Wang, an actual scientist and Princeton professor who sometimes answers our questions. Whatever the comments, facts ARE facts. Do stick around, NY Romneyite. This is a truly interesting, complex, earnestly fact-based site, dealing in the statistics of probability. I’m learning so much, and so is my husband who has a degree in physics.

    • wheelers cat

      This comment section is largely populated by blue phenotypes.
      Dr. Wang will welcome a red phenotype commenter, but you have to be intellectually honest.
      That is the price of admission.

    • NY Romneyite

      wheelers cat – sam pretty clearly stated that my points were relevant, outside of a slight disagreement about the flavor of 2008, which actually i agree with him on.

  • John M S

    Romney is winning now,…..the Math is out of breath, trying to catch up….

    • Sam Wang

      The math is fine, though it has to run in one direction, then another. A workout.

      State polls are fairly dense in OH and FL, so those are caught up – and show some return toward predebate conditions, especially OH (median Obama+4%, n=5).

      Agreed, Romney leads in national polls (median 1%, n=6) – but again there is bounceback by a few points from 10/4. National polls disagree with state polls by about 3%; I commented last week on the Obama campaign’s seeming swing state advantage, about 2% then.

  • Obama 2012

    I was thinking that we were close to a bottom from looking at the incoming polls… nice to see that bounce at the bottom indicating that I was right (hopefully)

  • Matt McIrvin

    I don’t think we’re at the bottom yet. Big tick down in the RAND poll today, apparently caused by a large decrease in likelihood of voting among the Obama supporters. I think the RAND sample is getting affected by the Obama-as-loser narrative in the media.

  • DaveM

    I dunno…this morning’s state polls look pretty good from Obama’s persepctive. I’m finding it takes a bit of self-reorientation to remove fresh numbers from the narrative they tend to be accompanied by (“NBC/WSJ/Marist poll: Romney gains in key swing states”) and examine them in the context of other current numbers from the same states, rather than previous numbers from the same pollster.

    If I’m understanding how the Meta-margin is calculated, seems like it should be ticking upward this AM…

  • Joel

    Sam, looks like you’ve made the bigtime, now that the trolls have started filtering in.

    • Sam Wang

      Correlates with Republican morale. The more reasoned comments are good, and refreshing in a way. We’ll see how long it lasts.

  • Froggy

    A week ago, just after the debate, I predicted that the MM was due for a tumble. Today, it’s poised for a rebound. If you look at the six states where the oldest poll considered in the MM is at least partially post-debate, five of them would look better for Obama if you dropped that oldest poll (or polls, where there is a tie on polling dates). Here are the numbers:

    MI: O+3 -> O+4.5
    PA: O+3 -> O+4
    OH: O+1 -> O+1 (no change)
    CO: R+1 -> Tie
    VA: R+1 -> O+3
    FL: R+0.5 -> O+2.5

    NV is also poised to make a move for Obama. NH, IA, and NC, on the other hand, each have only one fully post-debate poll, and probably will move toward Romney when the polling catches up.

  • Rose Pleis

    The snake oil salesmen did a good job of convincing people he had the product they need. The question is when will the buyers remorse set in? I certainly hope before we are all stuck with him. A super job of selling does not change the fact he is a lemon and a lemon is still a lemon. It always cost the buyer money to repair. I appreciate this site for the brainpower involved to keep things in perspective.

Leave a Comment