Princeton Election Consortium

A first draft of electoral history. Since 2004

Romney rallies the faithful

October 8th, 2012, 3:00pm by Sam Wang


Update, 5:05PM: The Meta-Margin just changed to Obama +2.3%. Maybe let it stabilize a bit more…tune in tomorrow. -Sam

The varied nature of state pollsters is often regarded as a minus compared with national polls. There are partisan outfits, little organizations, one-shot polls, and so on. However, one good thing: when partisan-leaning pollsters get excited, they might rush out and do a survey. This creates a quicker (though possibly overamplified) picture of changes when they occur.

Here is what they show: the post-debate bounce is looking a lot like the post-Ryan-VP bounce. It is about 3.0% in size, closing half the gap between Romney and Obama. As debate-induced changes go, this is exceptionally large. It’s a bit early to say for certain, but there are signs that Romney’s bounce is peaking.

We got a very fast read on the post-debate change: a 3.0% shift in the Meta-Margin, basically in one day. Thank you, Rasmussen, Purple Strategies, and We Ask America. That change is impressively large: it’s bigger than the 1-2% that appeared in 2004 and 2008. It is starting to level off now.

The effect of the debate is looking quite a lot like another event that energized the Republican base: the addition of Paul Ryan in the VP slot. That rise occurred over about a week, and then Todd Akin shot off his mouth. At that point, the race stayed mostly immobile until the Democratic convention got things moving back toward Obama.

Two bits of evidence suggest that Romney’s post-debate gains have come from inspiring partisan voters, as opposed to flipping nonpartisans. First, according to a recent PPP poll in Wisconsin, gains have come in the form of an increased enthusiasm among Republican voters – but not among independents. Second, the RAND survey, which tracks individual sentiment, does not show a massive wave of one-way mind-changing, which is what occurred after the Democratic convention.

If the Romney campaign can keep their supporters fired up, the Meta-Margin should settle where the race has been all season: near Obama +3%. This seems to be a “set point” where this year’s dynamics have naturally gravitated.

Alternately, the Meta-Margin could reverse direction as the focus shifts to the VP debate and home-stretch campaigning. Maybe Team Obama could buy Rep. Akin some national airtime?

One thing I think it won’t do is continue in Romney’s direction as rapidly as it has since last Thursday. Extrapolated, that leads to a 59-41% landslide victory for Romney on November 6th. Obviously, that will not happen. Now let’s watch the graph unfold.

Tags: 2012 Election · President

62 Comments so far ↓

  • Nadia Hassan

    Professor Wang, have you seen the Kaplan et. al paper on a “mean reversion model of campaign effects”

    http://www.stat.columbia.edu/~gelman/research/published/psq_4021.pdf

    It seems to be applicable here.

  • Tapen Sinha

    Gallup rolling average of tracking

    The first number in the bracket is month, second, the date

    Obama +4 (10/3),
    Obama +5 (10/4),
    Obama +3 (10/5),
    Obama +3 (10/6),
    Obama +5 (10/7).

    So, Romney comeback lasted two days of a huge explosion?

    Tapen

    • Sam Wang

      Those changes are too small to unwind, though you could try with this approach. The Meta-Analysis did extremely well at capturing the change within two days. Its accuracy is better than Gallup by a factor of 10.

    • Joel

      Pew has R+4 in their recent release. There is a huge amount of variability in the polls right now, to my eye. I do think the meta-margin here strikes it about right, as well as the more diverse EV outcomes that are currently showing up.

  • Jeff Smith

    Sam,

    Love this site!

    Two questions: First, following on Tapen’s post, in the Gallup seven day running average, Obama seems to have gone up two points over Romney today from yesterday. In order for that to happen on a seven day average, wouldn’t that mean that yesterday’s daily poll was 14 points higher than the one that dropped off from pre-debate? That seems a big shift (economic numbers?). I’m happy for any shift positive here, but want to check to see if I am doing this right.

    Second, the Realclearpolitics average of polls, which seems to be cited quite often, seems to me to have a right wing bias in it. It appears to be the case that Obama-favoring polls get dropped from the average sooner, and Romney-favoring polls stay on longer. Sometimes this takes the form of a +1 Obama poll being kept on when Obama is at +3 or so on the average. Has anyone else noticed this?

    • Brian

      I can’t easily find details on RCP’s calculation methodology but it’s possible that their “average” is a mean, rather than a median. Means will be thrown off by outliers, whereas medians will neutralize them. If there are more right-leaning outliers than left-leaning, this could tip their average rightwards.

    • Sam Wang

      Indeed, they use an average. I believe all the other aggregators do.

    • Iseeurfuture

      Jeff,
      You are correct in your statement. I’ve just visited RCP (in which I hate to do) the new Gallup Poll that was revised and released today had BO 50-45 but on RCP currently Gallup is listed with a tie and has an asterisk. Pew who came out with their Poll release has MR 49-45 but that’s PERFECTLY put in place. It’s why I come here. Sam always gets it right. Meanwhile DEMS’s are freaking out.

  • Prairie Pundit

    The current HuffPo headline — “Pew: Romney Up by 4!” — freaked me out a little, so I came running here. I suspect I am not alone among numerophobes in routinely seeking comfort in this Web site. And Professor Wang never fails to deliver! For the first time in my life, I find math soothing.

    • Jay

      The reason HP put that up there is to scare lazy democrats into voting. Its sort of a wakeup call.

    • Olav Grinde

      Jay: “The reason HP put that up there is to scare lazy democrats into voting. Its sort of a wakeup call.”

      It should also be noted that the Huffington Post has new ownership: AOL. I have noted some very marked changes there since Arianna Huffington sold out, but even more so in recent months. This is apparent in all sorts of ways, ranging from photo choices, headlines you wouldn’t see before, far more celebrity and other stuff devoid of real content…

    • Richard Vance

      Sam, I love your sight and agree totally with he logic. Given we still have the warped EV system them the EV is the ONLY thing that matters.

      I certainly hope this poll aberration will scare the BeJesus 0ut of all the dems. They were getting too cocky and the President slept through a debate(sic) where all he had to do in response was to hold out an Etch-A-Sketch and say “here yours looks used up.” Where are these strategy people and why is Obama so obviously ignorant on Etch-A-Sketch and other zinger opportunities? Its obvious that with Romney doing what 100 debates this year and Obama doing zero that they should have had used Chris Matthews as Romney in debate preps?? Instead they softballed him and look what happened. Even I was thinking Obama’s not fit.. If he can’t call a stooge a stooge what’s he doing in that big chair?

  • Arbitol

    Does the Pew poll give you pause, Sam? Have we seen the black swan event?

    • Prairie Pundit

      I’ve been wondering the same thing – can a debate be a Black Swan event? I mean a debate, such as last week’s, that fits within the standard expectations for such events, which allow for the possibility that one candidate will clearly outperform the other. It’s not as if the President wet himself on stage or strangled a kitten. Still, I wonder. . .

    • Sam Wang

      Certainly the debate was expected to move opinion. The question was how much. I am surprised.

      At the same time, several months ago I suggested the center point of the campaign was Obama +3.0 +/- 2.2. The race is currently 0.3 sigma below the midpoint. The lowest it’s gotten was right after the Ryan VP nomination — Obama +1.6, or 0.7 sigma below the midpoint. Like Chou En-lai said about evaluating the French Revolution…too early to tell.

    • Matt McIrvin

      What concerns me is the possibility that such stories can now create self-fulfilling prophecies at an unprecedented rate. It could be that Obama’s performance wouldn’t have been a deal-breaker in any previous election cycle, but now, with multiple self-reinforcing news cycles happening over the course of a single day, if you do this it creates a runaway feedback loop and the other guy wins in a 400-EV landslide.

  • L. Murray

    I have to agree with Prairie Pundit. To see such a big number for Romney in the Pew Poll is very alarming, but I feel somewhat better now. As a humanities person, a great deal of the mathematics here is a little daunting, but if the bottom line is that President Obama is still leading as the election approaches, then Maybe I will get some sleep tonight.

  • Steve in Colorado

    Would there be any info to be gained by calculating the area between the 2 lines on the RAND voter shift graph?
    This seems to be one of the best polls for showing actual momentum.

  • Ram

    When all the dust is settled, this race appears to be very stable. Mitt Romney has never crossed more than 47% in Gallup poll since Spring with 44% as his floor. On the other hand, Obama starts with 47% floor and has often reached 53%.

    Thus my take this appears to be an Obama’s election from 51%-49% to 52%-48% depending on how 1) eventful three debates turn out and 2) get out the vote efforts.

  • Arbitol

    And just like that, the Meta Margin is down another point. Pour me a tall one.

  • Mark F.

    If Romney ups his PV total by 2% , he might win a solid South with Ohio and CO putting him over 270. Democrats should be worried, especially if he has 2 more good debate performances.

    I am thinking nobody will get 50% of the popular vote and the election will come down to a very few states.

    Obama is still favored though. Ohio seems like a tough nut for Romney to crack and he really needs it.

    Anyone else sensing a really close election?

    • Matt

      Maybe, but according to Sam Obama has a clear and meaningful lead. What we’ve seen is Obama open up a bit of a lead up to four or five points and have it revert back to about two. This has happened several times. I suppose it could go all the way to tied within the the next month, but it would take another meaningful event like the first debate. I suspect Romney will need something beyond another couple debate wins given expectations and the fact that the next debates are likely to have diminishing returns even if he did well.

  • Howie Weiner

    As I said last week the entire Meta-Margin will be gone by the end of the week. The debate was a black swan event, immensely damaging to the President. I’m not saying that Obama will lose the election, and I sure hope not, but the political effect of the President’s ineffectiveness will outweigh the otherwise unchanged fundamentals of the race. The polls will revert to their mean over the next few weeks but only if there is not a duplication of the last debate. And that starts tomorrow night.

    • Joseph

      This is a manufactured “black swan” event, if it’s even that. That is to say, it was totally predictable that, no matter how badly Mr. Romney did, he was going to be cast as the “winner” by the right wing-controlled media. Having even a marginal “win” has been recast therefore into a monumental event. Consequently, it’s not anywhere near “immensely damaging” to the President. Yes, the Republicans have gotten excited about their candidate. Yes, some of the bloom left the rose for the President. But the basic dynamics of the race haven’t changed. Actually, one could say that one dynamic has changed, but that happens to be more positive for the President than negative, and that is the improvement in the jobs numbers. In the long run, I suspect that will have more of a lasting impact than the debate.

  • Barry

    Guys, take a breath.

    All day, we’ve been seeing the bump begin to subside, as Sam first indicated, and some polls trending back to O (Gallup 7-day).

    Pew has the race TIED UP at 46 each for REGISTERED VOTERS. The purported 4 point lead is after their likely voter screen.

    We already have indications that what Romney’s debate performance did was rev up his base. My take is that the even race for registereds is the top of the bump, covering 4 days after the debate. A likely voter screen is a judgement call as to “likely.”

    Watch the state polls, as Sam always says.

  • KEL

    What is going on with the polls? Why should they move at all in either direction? It would take an extreme event indeed to cause me to change my vote. And i suspect the partisans in the opposing camp feel the same way. The frames as George Lakeoff would label them are quite distinct.

    So what is happening in the minds of the persuadable. Or is what we are observing mearly noise in the polling data?

    • Matt McIrvin

      My best guess is, it’s low-information voters who would have been undecided in a “pox on both their houses” way, and maybe some third-party types who were thinking of voting for Gary Johnson, coming over to Romney, combined with Democrats getting dispirited and Republicans with low propensity to vote getting fired up. Probably not many outright Obama supporters switching to Romney.

  • Steve in Colorado

    I work in Boulder, CO, but many of the people here come from farther away. What I have seen here are 3 people (out of ~100, though I only talked to a couple dozen) switch from undecided or Gary Johnson to Mitt Romney after the debate. These are 3 people who consider themselves conservative but disliked Romney enough to go 3rd party or to not vote.
    However, the majority here still support Obama. Obviously, this close to Boulder County, it will not be the same across the state, but this area has grown vastly while the mountains have gained much less population. And even the Springs is showing signs of a slight demographic shift as well. So even with these bad numbers in the past 5 days, I actually have an optomistic feeling about Colorado as compared to where it was in July (when I felt it was leaning slightly to the GOP).
    This is all right leaning undecideds getting some enthusiasm, enthusiasm that we’ve been building since the Democratic Convention.

  • Andy

    Many thanks for the excellent analysis, Sam! One of the things that the meta-margin makes clear is that Romney has had a ceiling (so far). Obama has never fallen below a little under +2 in the mm. This reminds me of Romney’s problems in the primaries. He had a “closing-the-sale” problem in that he struggled to win over 40% of the vote in the early primaries (with the exception of Florida, where he spent some $15 m). Only when Sanotorum dropped out in April did Romney start winning majorities in the primaries. Romney seems to have had a similar ceiling in presidential polling so far. As Ram writes above, “Mitt Romney has never crossed more than 47% in a Gallup poll since Spring with 44% as his floor.” The same point is evident in the mm. One wonders if Romney’s ceiling is a consequence of widely discussed changes in electoral demographics. We may soon find out if that ceiling is real, especially if Obama (or Biden) perform poorly in the upcoming debates.

    Andy

  • chris brandow

    can you comment on whether the relative dearth of polling, cited by Nate Silver, is having any impact on the accuracy of the meta-margin? higher SDs?

  • A New Jersey Farmer

    Greetings all. Most of the polls released so far were completed at the height of, or a day after, Romney’s largest debate bounce. We’ve already seen some regression in Romney’s support and, as some have noted, a small rebound for the president. The state action is where it’s at. Romney could win FL, VA and OH, but as long as Obama holds on to WI, NH, CO,and IA, he wins reelection. Given the fundamentals and the mm, what are the probabilities that the president loses?

  • A New Jersey Farmer

    Exactly. And as other pollsters have noted, party ID can be fluid depending on such factors as who has momentum, such as Romney does right now.

  • Ram

    In Obama won, in 2008, by 52.9% with 365 EVs. The current fundamentals like jobless, GDP, demographics, predict this to be a closer race than than 2008. This race was never destined to be a similar landslide either way. After all the VP selection, convention, debate, 47% comment, etc related noise, the current race appears to be trending back to where it belonged to: A CLOSE RACE with demographics and EV collage advantage compensating for less than satisfactory economy.

    If Romney had not run such a lousy campaign, the polls probably would have shown a closer race than what we saw till this week. Now that he has moderated his positions, unexpectedly had one sided debate performance and is probably running a decent campaign (without any gaffe in the last two weeks), the race is attaining its equilibrium.

    Sam is probably right with Obama prevailing with around 290 EV to 332 EV depending on how the campaigns run the last three weeks but for a extraordinary event or campaign mistake.

  • Ed Groome

    The “With +2% for Romney” map gives him the electoral edge for the first time in a long time. Perhaps the Presidential election is back in play after all and worth our notice? It was mere days ago that we were told there was little point in paying attention to that race, and that could prove out, but it sure seems scary now!

  • Olav Grinde

    I fear a Black Swan event on election day itself, and it is this: that the net sum of the voter suppression efforts will be far greater than we expect — especially in the swing states.

    We may be very surprised, indeed, at the difference between the polls and counted votes.

    • wheelers cat

      I think Dr. Wangs analysis on redistricting is sound Olav. It wont affect the presidential, but it can affect downticket races.

      I prefer this RAND graph.
      https://mmicdata.rand.org/alp/?page=election
      in the past 24 hours there has been only .09% improvement in Romney’s position.
      The shift graph is misleading I think. Romney started in the hole.

    • Matt McIrvin

      RAND actually is starting to show something I expected, namely a change in who the respondents expect will win. But I thought it would be much more drastic.

    • Matt McIrvin

      …The other thing to remember about the RAND poll is that it ought to be used for eyeballing changes, not absolute positions. I think the reason the gray bar in their election forecast is so huge is that they don’t know how representative of the electorate their fixed sample is; in a more conventional tracking poll, you reduce that uncertainty by sampling the whole population repeatedly.

      To figure out the overall state of the race, it’s better to go with state-poll aggregators like PEC, 538, electoral-vote.com and Votamatic. But right now they’re all hobbled by the fact that we have a dramatic delta-function event developing during the temporal spread of the aggregation, and even during some of the individual polls within the aggregate.

      I suspect that most of them are currently showing a better situation for Obama than is the case. But a few days from now they’ll probably be showing a worse situation for Obama than is the case. Temporal sampling effects abound.

  • DaveM

    A New Jersey Farmer says, “party ID can be fluid depending on such factors as who has momentum, such as Romney does right now.”

    It would be interesting to see what kind of relationship there might be between significant events during the campaign and swings in party ID among pollsters who don’t weight for that. In other words, are most “bounces” mostly a result of shifts in party ID composition of polls following such events–presumably due to a greater response rate to pollsters among those whose enthusiasm has been uplifted?

    • xian

      no, it’s more than that. one way people express enthusiasm for a ticket is by identifying with the party.

    • Matt McIrvin

      Weighting by party ID is generally a bad idea, because for many people party ID is basically a proxy for who they support in the presidential race.

      I think Rasmussen is the only major pollster who does it. If I recall correctly, they occasionally readjust their model of the party makeup of the electorate, which makes their numbers jump; the rest of the time, nailing the results to a static model of party ID tends to suppress changes in an unnatural way.

  • Job

    Dr. Wang,

    I’m interested in the effect of the motivated base. Let’s imagine a scenario in which Ryan trounces Biden, and the GOP momentum carries through the following two presidential debates. Few minds are changed from the opposition, so it becomes solely a question of who can get the base out.

    What sort of turnout numbers would the GOP need to see to make a Romney win happen? And what sort of turnout would the Democrats need to stop this? Is this statistically probable?

    • wheelers cat

      Job, Brownstein, Medved and other pundits have calculated that if Obama has equivalent turnout to 2008, that Romney would need 63-65% of the white vote to win.
      That seems pretty impossible given that 33% of whites vote liberal, even Carter got 33% of the white vote.
      And Reagan only got 60% at the top of his game.
      There just may not be enough whites left in the electorate to carry the election for the GOP anymore.
      IPOF, Kerry would beaten Romney this cycle. He got 37% of white males and 41% of white females.
      ;)

  • Olav Grinde

    I am wondering to what degree some presidential polls might be off, if they don’t include third-party candidates.

  • AlexP

    Hi Sam! First time posting here–really love your site. It’s truly a breath of fresh (math) air, especially to a former mathematician from Princeton.

    A pair of related questions to you.

    1. Do you see that historically there is a tendency for the final election result to be tighter than the final polls
    2. Do you think that polls tend to magnify the momentum of the day, i.e., those supporting whoever seems to be on the rise (not necessarily the one leading in the poll) more likely to respond and be open to the pollsters?

    In a way, if 2 would explain 1 if it’s an accepted phenomenon in psychology.

    Would love your thoughts.

  • Venkat Ranganathan

    What is the reason for another point of downward revision in the meta-margin? I saw that the meta margin was stabilizing. Is it after the foreign policy speech by Romney? Thanks

    • Matt

      A few more state polls dropped that showed a closer race in MI among others. It responds only to state polls and updates several times a day.

  • Ed Groome

    Party ID can be fluid, as can everything else. At the end of the day, PEC and anyone else trying their statistical hand here is in the business of predicting the future, specifically the future behavior of human beings, too many of whom don’t know if or how they will vote when it comes to actually doing it. It’s an entertaining exercise, no doubt, but a fruitless one, I fear. That certainty can change to uncertainty so suddenly should come as no surprise to anyone who has lived for a few decades.

  • pechmerle

    Olav, I don’t think you’ll see as much of an effect from GOP vote suppression efforts on election day as you fear. Successful court challenges have substantially eliminated or mitigated those GOP efforts, for this cycle.

    Mitigation has partly meant court rulings that the i.d. etc. requirements were brought in too close to election day to give voters a proper chance to react to and comply with them. Also too brief a period to properly train poll workers and administer the changes.

    That kind of mitigation court ruling, though, does have an implication for 2014 and beyond: some of these same courts would allow these suppressive rules to be enforced after such a longer lead time.

    And then you get to an interesting dynamic. GOP voter suppression efforts will continue with a goal of offsetting their increasing demographic disadvantages. Ultimately, I would expect that the GOP has to shift some of its policy positions to remain competitive. Indeed, we know perfectly well that Romney understands that (it’s how he became governor in Mass.) but couldn’t win his primaries this year if he acknowledged or adapted to it.

  • Iseeurfuture

    Nate Silver’s analysis of the recent PEW poll and MR’s bounce put into perspective. He’s finally realizing the hype or at least somewhat calling it out as it should be.

  • wheelers cat

    I suffer from night terrors and brutal insomnia, and I just woke up in a terror sweat from a dream of President Romney to find RAND still essentially and reassuringly unchanged.
    https://mmicdata.rand.org/alp/?page=election
    Nate is also very calming to read. Usually his high verbal equivocations act on me as a deadly soporific and this is no exception.
    “blah blah blah….but the gains that he made on Monday in particular were all because of a single poll.”
    I think the Pew results could easily be explained by increased enthusiasm in the GOP leading to an increase in response. Consider the Palin phenomenon…her addition to the ticket actually flipped the curves temporarily, just on the basis of rabid base enthusiasm.
    “Pechemerle: Ultimately, I would expect that the GOP has to shift some of its policy positions to remain competitive.”
    Exactly. Again, I believe that there are just not enough white conservative voters left in the electorate to deliver the 63-65% of whites necessary for Romney to win. The GOP has to adapt or go extinct.
    The death throes of an organism are never pretty.
    I think waiting to see how the poll aggregates are effected this week is advisable.

    O lente lente currite noctis equi

  • Bill N

    The rapid decline in the meta-margin has been, quite frankly, rather shocking. My understanding is that Gallup is about to release a new poll that is for the first time using a LV as opposed to RV model. The expectation is that Obama’s lead in this poll will be less than it has been with the RV model. Also, I think a Kos poll is about to be released showing Romney leading. My guess is that the meta-margin over the next few days is going to drop below 1%, and maybe below zero. Sam, is it about to reach the point where the Presidential race is now knife edge?

    I am still stunned that a single debate could apparently have this kind of effect. If it is the debate, it suggests to me that all you have to do to win a Presidential election now is be obnoxious during a debate and throw out as many lies as possible using the Gish Gallop debate technique. Sad.

  • Olav Grinde

    Wheeler’s Cat & Iseeurfuture: It was indeed heartening to see the very different image that emerged when Nate Silver at FiveThirtyEight.com put the Pew Research poll into perspective. Very in much in line with what Dr Wang has been saying for some time…

    By the way, there does seem to be significant new development: Mr Romney has discontinued his series of non-stop gaffes. Perhaps his handlers have tightened the reins. Romney’s recent foreign policy speech was careful cherry-picking, indeed, with hardly any indication of anything that he would do concretely different from President Obama.

    Must confess I’m holding my breath here, and am especially watching the downticket races. Won’t be long until the Biden–Ryan debate…

    PS Shouldn’t we soon be seeing enough polls on Congressional races to be able to make outcome predictions based on actual polls?

    In the meantime, I am fascinated, and rather dismayed, by Dr Wang’s revised House analysis, afte he revisited gerrymandering and other factors. It does not surprise me that the GOP has “reaped the rewards” of Karl Rove’s strategy: winning control of as many swing states as possible in 2010 — and then changing the rules of the game.

    • wheelers cat

      Like Nate says, its a single poll. As for Rove’s “strategy”– it isn’t a strategy. It is a tactic in the conservative war against demographic evolution. Its a delaying tactic. It is profoundly anti-democratic, but it can only postpone the inevitable outcome. Game Theory 101 strategy beats tactics in the long game.
      And yes, Romney’s debate performance is what my hero, Nassim Nicholas Taleb would call a black swan event. It is a negative BSE for our side, a positive BSE for theirs.
      Has no one but Ed Freeland and me read his books?

  • Matt McIrvin

    My current best guess is that Romney had really good days on Thursday and Friday. Maybe cataclysmically good: Pew’s R+4 result may be exaggerated, but PPP apparently did a survey for Kos that also showed him in the national lead during that time, whereas early indications of later results seem to show better numbers for Obama.

    We’re still seeing new poll numbers coming out that cover the Thursday-Friday period, and even if Obama had an impressive recovery over the weekend, the numbers are going to keep looking worse for him until polls whose sampling window was in that time period dominate the aggregates.

  • Michael Worley

    Sam,

    1. Any chance you could start doing a 3pm update?

    2. Is a drop from 2.0 to 1.84 in the MM statistically significant?

  • Steve in Colorado

    We’re going to see about another week of poll numbers with this debate ‘bounce’ in them. Until then, I’d recommend putting on your Hitchhiker’s Guide to the Galaxy ‘Don’t Panic’ sunglasses.
    Unless you want to start doing to Obama what the right-wing bloggers started doing to Romney after the 47% video came out.

  • Steve in Colorado

    I do have a question about the ‘Power of your Vote’ chart. It would seem that the closest races and the states with the most electoral votes should have the most power, but it doesn’t seem to be that way, as an IA voter seems to have a much less competitive race and fewer EV, yet has about twice the power of a CO voter.
    Can anyone explain this to me?

  • A New Jersey Farmer

    Notice that even with all of the polls released today, the mm has stayed relatively stable as of 8:05pm.

  • pechmerle

    This is — from The Atlantic and featuring the insights of Jim Fallows, who is well respected by Sam (and me) — is well worth reading on why Obama performed so poorly in the first debate.
    http://www.theatlantic.com/politics/archive/2012/10/a-limited-and-self-serving-defense-of-political-punditry/263338/

    The pundits who said Obama didn’t seem like he even wanted to be there appear to have been exactly right. Also excellent thoughts on why a sitting president doesn’t have the time and inclination to do proper debate prep.

    But Fallows makes the excellent point that Obama is a highly competitive guy (the way he sharpened his tone in the following couple of days also illustrates that), and he will want to show people and ‘that darned Romney’ that he is not really to be trifled with. We shall see.

    Fallows makes many excellent points, which I commend to you.