Update to my prediction that Romney will win, with an interesting new forecasting model

Summary:  In February I predicted the outcome of the election:  President Romney will prove an effective President, reshaping America for his key supporters, 17 April 2012. Here’s an update.

Graphic from SmartMoney, 22 April 2011

The polls show a tight race (see Nate Silver’s blog at the New York Times). With so few undecided voters, which party gets the greatest turnout might determine the winner. Especially since so few Americans vote.  Heavy rain in Ohio might change US history.

Let’s look at a different methodology.  There has been much media attention to the computer model by Ken Bickers and Michael Berry, Professors of Political Science at the University of Colorado – Denver. It successfully hindcast all 8 presidential elections back to 1980 (the earliest for which  economic data is available).  Here is their latest report:  “Forecasting the 2012 Presidential Election with State-Level Economic Indicators” in Political Science & Politics, October 2012.

Their forecast:  a massive win for Romney:

“Our prediction, based on the model analyzing returns from the prior eight presidential elections is that the president will win 17 states, plus the District of Columbia. … our forecast is that Obama will receive 47.14% of the two-party popular vote. Using confidence intervals around each individual state forecast and aggregating to a national popular vote as earlier described, our model projects a likelihood of 77% that Romney will receive a majority of the ballots cast for the two major parties.”

Abstract

.

From Berry & Bickers (Oct 2012)

Nearly all forecast models of US presidential elections provide estimates of the national two-party vote (Campbell 2008). Each of the 9 forecasts published in the 2008 forecasting issue of PS: Political Science and Politics made national popular vote total predictions for the major party candidates, while only one provided an expected result in the Electoral College (Klarner 2008).

These national vote models are assumed to be reliable forecasts of who is likely to win the general election. In most cases, this assumption is reasonable.

It becomes problematic, however, at precisely the point that forecasts are most interesting: when elections are close. In tight elections, national forecasts can and have produced a “winner” different from the actual winner.

Consider the forecasts and ultimate outcome of the 2000 election. Each of the 2000 presidential election forecasts predicted vice president Al Gore to win a majority of the two-party popular vote, which he did, but none correctly predicted governor George W. Bush to assume the presidency (Campbell 2001).

Never in US history have White House residents been determined through a national popular vote. Presidential elections are decided through contests in the states and the District of Columbia. The forecast model we developed explicitly models the presidential contest based on factors inherent to these 51 jurisdictions. This modeling approach allows us to make a projection of the Electoral College result, which popular vote estimates cannot.

For More Information

For more see the posts listed on these related FM Reference Pages:

Other posts about our elections:

  1. Lilliput or America – who has a better way to choose its leaders?, 19 November 2008
  2. About campaigns for high office in America – we always expect a better result from the same process, 17 June 2009
  3. The economy speaks – and votes, 12 August 2010
  4. The winners and losers from this election, hidden amidst the noise, 3 November 2010
  5. Why do awesome people – like us – have such inadequate leaders?, 2 April 2012 — Because we vote.
  6. The presidential debates are performance art. They’re Kabuki., 4 October 2012
  7. A reminder that debates are fun, not politics: Reagan had Alzheimer’s in 1984 and we didn’t notice., 5 October 2012

About the 2012 campaign:

  1. Campaign finance reform = incumbent protection, 20 December 2009
  2. Republicans have found a sure-fire path to victory in the November elections, 5 February 2010
  3. Mitt Romney and the Empire of Hubris. Setting America on a path to decline., 10 October 2011
  4. The hidden dynamics of the 2012 campaign, and what it’s doing to America, 9 March 2012
  5. President Romney will prove an effective President, reshaping America for his constituents., 17 April 2012
  6. Romney back on top. More evidence that the campaign news matters little. It’s the economy!, 19 July 2012
  7. The significance for America of Romney’s choice of Ryan as VP, 11 August 2012
  8. Our choice of a leader reflects our true self. What does 2012 tell about America?, 19 September 2012

.

12 thoughts on “Update to my prediction that Romney will win, with an interesting new forecasting model

  1. Romney is such an awfully flawed candidate that I don’t think this pattern detection is going to save the day.

    I used to gamble a bit on sports in my youth, back before the internet made data was so abundantly available. I was sitting at the Minneapolis Public Library going through old newspapers, looking for exploitable patterns. I found one where, for over 2 seasons, the Boston Celtics, when opening a multigame road trip the game always went low on the total score over/under. It was a pattern that held in almost 20 straight instances. So I started betting it. I opened with 2 wins and was feeling pretty smart. After that? 8 straight losses before I gave up on it.

    I hope I’m wrong, but this election prediction seems similarly flawed.

  2. Fascinating little article but Electoral College vs. Popular is not news.

    In my long voting life, I can’t recall a more dangerously inane, stupid presentation (and punditry) by both Parties. IF Romney wins, it will say that America has swallowed the worst of all the massive faux Econ Propaganda and the commensurate nonsense about Free Markets and self fulfillment. IF Obama wins it will tell me, Americans are still dangerously clueless (yet hopeful, wishful and not completely self-destructive….give them one more dose of disappointment.)

    “Heavy rain in Ohio might change US history.”

    This election is not about a change in American History….that has been happening for many, many years. This Place is not the Country its Myths try to portray and many of its citizens rely upon, in spite of reality. It is simply more of the same pretense we have lived with since the end of the Viet Nam War. The USA went to sleep then, shocked and childishly cowed and still wishes to remain in a dream state. The Country has been systematically stripped of most esential elements of a functioning Republic and the inhabitants are without a meaningful response….too many of us simply enjoy/rely upon the fruits of this undermining or are too frightened to accept the reality and face the consequences, thereof.

    The willful deceit by the dominant culture is almost breathtaking! Fascinating to watch.

    1. “This election is not about a change in American History – that has been happening for many, many years.”

      Life is a process of change, but Breton’s reply misses the point.

      If your Dad says that your marriage to Miss X will “change your life”, a son’s reply of “change has been happening for many years” is just sad.

  3. http://journals.cambridge.org/action/displayFulltext?type=6&fid=8700620&jid=PSC&volumeId=45&issueId=04&aid=8700619&bodyId=&membershipNumber=&societyETOCSession=&fulltextType=BT&fileId=S1049096512000984

    This “prediction” is entirely scientifically based upon data collection & scientific analysis. This is the same exact scientific model they used to predict Obama’s win in 2008, and all other presidential elections dating back to 1980. This is a non-paritsan analysis.

    1. JJJ — Why are you citing the same article that this post highlights?

      Also, they did not predict elections back to 1980. They “hindcast” them. Most models are tuned to successfully predict the past, so hindcasting provides only weak evidence. It would be more convincing if they predicted the result with out of sample data (ie, data for a period not used to tune the model).

  4. I think Breton’s post is exactly on point. Breton remarks: “If Romney wins, it will say that America has swallowed the worst of all the massive faux Econ Propaganda and the commensurate nonsense about Free Markets and self fulfillment. IF Obama wins it will tell me, Americans are still dangerously clueless (yet hopeful, wishful and not completely self-destructive….give them one more dose of disappointment.)”

    Precisely correct. This looks to me like a game of good cop, bad cop. Romney is the bad cop — ooohhh! Scary! Scary! Run to the good cop Obama and do what he says!

    Unnoticed? Both candidates promise to massively increase America’s already insanely bloated military/police/surveillance/torture state, both candidates promise to extend the crazy and unsustainable Bush tax cuts for the superrich, both candidates have abandoned all but lip service to wean America off its chronic oil addiction, both candidates have promised to shovel more trillions of dollars into the gaping maw of America’s catastrophically broken private for-profit medical-industrial system, both candidates promise to let America’s decaying infrastructure collapse further in order to build more Buck Rogers superweapons that don’t work like the failed but ongoing anti-ballistic missile system and the looney EAGLE system (a blimp 25 times the size of the Goodyear blimp with mirrors mounted on the underside that would reflect laser death rays sent from the ground) or the Rods From God scheme (orbiting tungsten telephone poles that can be used to bomb any location on earth), both candidates have swaggered and strutted with demented glee to threaten Iran with invasion and bombing and all manner of aggression even though Iran has done absolutely nothing anyone can discern against America…

    A friend of mine says that Romney is a deliberate plant by America’s elites designed to the lost the election so that good-cop Obama can offer us a kindler gentler path to abandoning the rule of law and throwing out the constitution and kidnapping and murdering American citizens without a trial and aggressively attacking foreign countries for no perceptible reason and extending unlimited power to corporations in the wake of the Citizens United decision. That sounds extreme to me, but I’m starting to wonder if he isn’t right.

    In any case Breton is saying something that seems important to me, and needs to be said.

  5. there is just no way Romney wins. He is not popular enough with key voting blocs. One thing for sure, polls and those with produce and project voting algorithmic predictions are in the business of being read and believed as relevant. At least 46% of the people will latch on to a poll regardless of what it says. In my opinion, the polls which DISAGREE with me are the ones I read the most.

    Nothing else in play here but the desire to be interesting and read. The election will be unpredictable up until 9pm eastern time on Nov. 6. Stop selling me model predictions.

  6. Thanks FM. Viking Death March reminds me of watching American swirling around in a toilet bowl. Down on your knees you don’t look so tall. Open your eyes and the empire falls.

    Sun Tzu would say the American empire is farce. I say it’s a butt f’ing farce.

    1. “watching American swirling around in a toilet bowl”

      Exactly my feeling often as I read the daily news.

      “American empire is a farce”

      I don’t know if that well describes it. Certainly not to those killed to expand it — and to those who die in its name. It’s a mad empire by the standard metric: its unprofitable.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.