Princeton Election Consortium

Innovations in democracy since 2004

Outcome: Biden 306 EV (D+1.2% from toss-up), Senate 50 D (D+1.0%)
Nov 3 polls: Biden 342 EV (D+5.3%), Senate 50-55 D (D+3.9%), House control D+4.6%
Moneyball states: President AZ NE-2 NV, Senate MT ME AK, Legislatures KS TX NC

Hurricane Ryan passes. Duration: one news cycle

August 24th, 2012, 11:59pm by Sam Wang

Take one last look at the Ryan bounce. I’d say it’s reached its peak. Whatever comes next will be obliterated by the transient effects of the two upcoming conventions.Ryan bounce: about 3.0-3.5% in Meta-margin units
Think of this graph as storm tracking. Hurricane Ryan has passed just in time for us to get a good look at Tropical Storms RNC and DNC. Explanation:

The Meta-margin basically means the same thing as the usual margin in national polls. It is defined as how much swing in state polls would be needed to create an electoral near-tie. It is accurate to <0.5%.

And here it is in our favorite units, electoral votes:
Ryan August 2012 bounce in electoral vote units

For each date, the quantity plotted is that day’s closing Meta-analysis. Polls are released at least one day after they are completed, so there’s a bit of a time lag in the graph. Even so, the effect of adding Paul Ryan (admiring his eyes and budget, temporarily stilling the drumbeat on Romney’s taxes, and so on) is clear.

Notes: The vertical grid lines are Sundays, indicating media cycles. To indicate the August 11th Ryan VP nomination I used “v” to indicate it as a likely race-shifting event. For the Akin statement on “legitimate rape” I used “*” because all we really know is that media attention shifted, not necessarily that Akin’s horrific and false statement affected the Presidential race.

In my view, what happened here is that Ryan had his week. And as we know, our diverse panoply of mass media can only pay attention to one thing at a time.

To new readers: welcome. If anything in a post is ever hard to understand, sorry – I am writing cumulatively as the campaign season progresses. Make sure to read comments. Many of the commenters ask probing questions and make excellent points. (Update: for example, see this one today.)

Want to play with the Meta-analysis output yourself? The core daily output is here. Lots of code-y goodness here.

Tags: 2012 Election · President

13 Comments so far ↓

  • Olav Grinde

    That’s a fascinating analysis, Dr Wang!

    I think one thing that is highly problematic for the American public discourse, and thus for American democracy, is the mono-manic character of the news media. By that I mean the news media’s inability to really focus on more than one main story at a time.

    Time and again we see how a story disappears from the headlines, not because it is resolved, but merely because it is displaced. And thus, instead of having a real fight about substance, we have a skirmish of tactics that are designed to define the dominant story.

    The result is a myopic media and a myopic electorate, with the focus shifting in a haphazard manner. How in the world can you have a meaningful public discourse under such circumstances? Well, the answer is that you can’t! (And you’re “not supposed to”.)

    A study of, for instance, the three-four top headlines of Fox News over time would be a highly instructive case in point. Or for that matter, you can take the headlines of CNN, NBC, etc…

    Perhaps the core problem is the corporate nature of the news media, where it is a question of ratings — news coverage thus becomes merely a branch of the entertainment industry.

    The decisive role of investigative reporters is, tragically, a thing of the past. What we have instead is photogenic anchor men and photogenic anchor women, and ditto reporters, with teleprompters. (Ask yourself, for a moment, the average age of 60 Minute’s journalists… They’re septaguenarian!)

    Instead, the most “trusted” names in news are Jon Stewart and Stephen Colbert. Entertainers! Really penetrating questions are left to comedians!

    Where are the critical, uncompromising questions? Where is the real — broad and in depth — news coverage? It’s missing, and so is the corresponding public debate that SHOULD be taking place in advance of the coming national elections in the United States of America.

    These circumstances are truly tragic — and their consequence far-reaching, beyond imagination.

  • Tapen Sinha

    Talking about the “hurricane”, if I may, I would like to tell you a (true) story. Back in 1999, I had just finished my book taking a good look at pension privatization in Latin America (that was published by Kluwer in their Huebner international series on risk, insurance, and economic security in the following year). A friend of mine from the University Wisconsin (where I used to work) asked me if I would like to debate the issue of pension privatization.

    He mentioned a newly minted Congressman from that district has been making a lot of noise about how wonderful privatization of social security is and so on. So, he organized a debate between the Congressman and me as an event. I asked my friend to send me all the stuff that the congressman had written about it. He Fedexed me a big stack of papers. I went prepared. I asked them to let him speak first and I would go last.

    So, the guy went first blabbering about what wonders privatization of social security done in Latin America. His talk was total all fluff – no evidence – just opinions. I talked about all the issues carefully backed up with data.

    “It was a knockout!” My buddy told me later. I managed holes in every single argument this guy had. I ended by saying, “Evidently, he neglected to take proper economics of social security classes at Miami University of Ohio.” He got so red in the face, I thought he was going to get up and punch me.

    That was Hurricane Ryan.


  • BillSct

    Truth through humor has always been the role of the Court Jester; look at Shakespeare. The nation needs Stewart and Colbert.

  • BillSct

    For another Jester’s point of view, take a look at Bill Maher’s piece on magical thinking, “My New Rule for Todd Akin and the Republican Party”

  • Ottovbvs

    Apropos the comments of OlavGrinde, I remember going to see a play in London in the late 60’s the central premise of which was that in future society would be divided into two classes. A super rich elite of beautiful people and the mass of proles who would be kept docile by watching endless TV programs featuring the elite having sex in all kinds of different ways. Given societal income shifts and the ubiquity of reality TV has the premise of this play arrived?

    Jefferson on the role of ridicule in exposing nonsense:
    “Ridicule is the only weapon which can be used against unintelligible propositions. Ideas must be distinct before reason can act upon them…..”
    He was talking about religion but it’s a sound general principle.

  • wheelers cat

    Olav the JIT(just-in-time) news cycle is the new normal.
    This is basically because of SNT(social network theory), the flattening of of information provided by the internet, and the “freed” market of information.
    Media has simply been subjected to regulatory capture by its audience.
    People wont pay to read or view things they disagree with, so you can view the current state of media as buffeted by market forces.
    The rise of FOXnews is part of this phenomena.
    Here is a quote from Julian Assange about part of this dynamic.
    Moderator: The question has to do with the shift, alleged shift at Wikileaks from simply posting the material, having it crowdsourced, and people interpreting it, to actually interpreting what it means. Is that a change?

    Julian Assange: No. That’s part of the right-wing reality distortion field (some laughs in audience). Mother Jones has had some changes in the past few years.

    No, there hasn’t been a change, whatsoever. Although of course it was our hope that, initially, that because we had vastly more material than we could possibly go through, if we just put it out there, people would summarize it themselves. That very interestingly didn’t happen. Quite an extraordinary thing.

    Our initial idea — which never got implemented — our initial idea was that, look at all those people editing Wikipedia. Look at all the junk that they’re working on. Surely, if you give them a fresh classified document about the human rights atrocities in Falluja, that the rest of the world has not seen before, that, you know, that’s a secret document, surely all those people that are busy working on articles about history and mathematics and so on, and all those bloggers that are busy pontificating about the abuses in Iraq and Afghanistan and other countries and other human rights disasters, who are complaining that they can only respond to the NY Times, because they don’t have sources of their own, surely those people will step forward, given fresh source material and do something.

    No. It’s all bullshit. It’s ALL bullshit. In fact, people write about things, in general (if it’s not part of their career) because they want to display their values to their peers, who are already in the same group. Actually, they don’t give a fuck about the material. That’s the reality.

    So, very early on, we understood from experiences like this, that we would have to at least give summaries of the material we were releasing — at least summaries — to get people to pick it up, to get journalists to pick it up to get them to dig deeper. And if we didn’t have summaries to give a piece context, it would just fall into the gutter and never be seen again.

    This was about the failure of the crowdsourcing on the Afghan doc drop.
    So, what Assange did for the Iraq doc drop if you recall, was pre-release the material to media outlets to make stories out of it, and then flood the news cycle with a one day release. Al-Jazeerha did the best job (imho).
    But did you know, that the al-Jazeerha release was part of the reason the US got shown the door on Iraq? Muqtada al-Sadr was able to leverage the doc drop to get over 2 million sigs on a petition to expell the US, and use the stories to inflame populist sentiment to prevent Maliki from making a deal with the US to keep troops there.
    Of course, you wont see any stories on that here.
    People wont pay for it.

    We are experiencing regulatory capture in the “freed” information market is all.

  • Olav Grinde

    @BillSct: That’s a brilliant piece by Bill Maher. My wife pointed it out to me yesterday.

    How come Bill Maher is not doing his piercing commentary on prime time, say, on CBS? America needs clear-headed analysts like him, and serious investigate reporters like Dan Rather. Oh, wait…that’s right, they were on CBS.

  • Olav Grinde

    @Tapen Sinha: Thank you so much for sharing that great story about Paul Ryan! It deserves to be far better known.

  • Bill N

    I don’t know if you have done this, but an interesting idea occurred to me. It would be interesting and fun (yes, I am a data analysis nerd and junkie) to take your data from 2004 and 2008 and, using the data up to this point in time during those election cycles, make the same form of prediction about the outcome of the election that you have about the present one (i.e., your prediction that Obama has an 88% probability of re-election). Predict Bush’s probability of re-election, along with confidence intervals, for 2004; and Obama’s probability of being elected in 2008 along with confidence intervals. This would be a form of validity analysis for your current prediction. If you have already done this and posted the results, I missed it and could you please provide a link to that post? If not, could you do this analysis and post it? It would be very interesting to see.

  • Bill N

    I kind of thought you might have done this already. If my intuition about you is correct, you are a data analysis nerd and devotee just as I am (and that is meant as a compliment).

  • Olav Grinde

    Dr Wang, do you have any thoughts on the “likely voter” filter that many polls are now using? I’m wondering about the various adjustments that pollsters make, and your take on the validity (and variation) of their assumptions.

    Also, should there be a distinction between “registered voters” and “likely voters” polls? Are we comparing apples and oranges? Now, U suppose the might not matter much for Meta-Analysis… However, with regard to single poll(sters), might this give the impression of “movement” if the pollsters / newscasters fail to cast light on their changed methodology?

  • Olav Grinde

    * Now, I suppose that might…

Leave a Comment