The Keys to the White House by Allan Lichtman and Vladimir Keilis-Borok is a simple mathematical model that predicts who win a Presidential election. This model predicts who will win months or even years before an election. Let’s look at why Obama will win one year from now.
- Party Mandate: After the midterm elections, the incumbent party holds more seats in the U.S. House of Representatives than after the previous midterm elections. (FALSE – Democrats when from 203 to 193)
- Contest: There is no serious contest for the incumbent party nomination. (TRUE)
- Incumbency: The incumbent party candidate is the sitting president. (TRUE)
- Third party: There is no significant third party or independent campaign. (TRUE)
- Short term economy: The economy is not in recession during the election campaign. (FALSE)
- Long term economy: Real per capita economic growth during the term equals or exceeds mean growth during the previous two terms. (FALSE)
- Policy change: The incumbent administration effects major changes in national policy. (TRUE)
- Social unrest: There is no sustained social unrest during the term. (TRUE)
- Scandal: The incumbent administration is untainted by major scandal. (TRUE)
- Foreign/military failure: The incumbent administration suffers no major failure in foreign or military affairs. (TRUE)
- Foreign/military success: The incumbent administration achieves a major success in foreign or military affairs. (FALSE, unless you count Osama Bin Laden’s death)
- Incumbent charisma: The incumbent party candidate is charismatic or a national hero. (TRUE)
- Challenger charisma: The challenging party candidate is not charismatic or a national hero. (TRUE, assuming that Romney or any of the GOP front runners runs against Obama)
There are four “Falses.” When five or fewer statements are false, the incumbent party wins. When six or more are false, the challenging party wins. It looks like no matter who the Republicans nominate (see #13), five or fewer statements will be false. If the Republications nominate a war general or someone truly charismatic (say, Sarah Palin) and there is a viable third party candidate (like Ross Perot, not Ralph Nader), then perhaps Obama can be defeated.
Some would claim that Obama has lost his charismatic charm that won us over in 2004 (and again in 2008). If that is the case, then the Republicans would still need to nominate a war general or someone truly charismatic. None of the Republican frontrunners fit the bill.
Other blog posts on Presidential elections:
Punk Rock Operations Research 
November 8th, 2011 at 7:04 am
The money says its a coin flip
Barack Obama to be re-elected President in 2012
49.8%
http://intrade.com/v4/markets/contract/?contractId=743474
November 8th, 2011 at 8:47 am
8. and 9. are interesting to me. I would define the rise of the Tea Party as social unrest. Also the Fast and Furious weapons program could be considered a scandal. These two points seem pretty ambiguous to me and can point either way. I’m with Iamreddave and think its too close to call.
November 8th, 2011 at 9:10 am
I think it’s too hard to say yet, too. But on the other hand, having some sense of what could occur one year out is useful. Anything could happen, but the deck seems to be biased in favor of Obama at this point (that in itself is amazing!)
The model implicitly assumes that all 13 of these issues carry the same weight. It might have worked out in practice during previous elections, but it seems like the economic issues carry more weight this time around. We shall see.
November 8th, 2011 at 9:12 am
The estimable Nate Silver discussed this question a while back here and here. Definitely worth a look relative to this discussion.
November 8th, 2011 at 9:14 am
You beat me to the punch, Matthew! I was planning to blog about Nate Silver’s model, but it didn’t seem to fit into this blog post. I agree–it is worth a look. However, the methodology he used (forecasting algorithms) seem to be a bit dubious in that they rely on approval polls and economic indicators and make implicit assumptions about what matters (and in what proportion) to voters.
November 8th, 2011 at 12:33 pm
What concerns me about this model is that many of the questions are subjective and ambiguous. 7: does Obama really effect changes in national policy following the 2010 midterm election? 8: does the Tea Party – or even Occupy – represent a widespread unrest or simply the viewpoints of a vocal minority? 11: Will Obama be associated with success for capturing Bin Laden and the regime changes across the Arab world? I suspect that different people will answer these questions differently, and these are enough to change the outcome of the model.
Like others have said, it’s not conclusive right now, and there’s a lot of time until the election for things to change.
November 8th, 2011 at 4:44 pm
Greg, I know what you mean, but the authors of the study define them in more detail (they wrote a book about it!). Sure, you can define things differently, but then they model would not yield the correct results for past elections, so it would be useless. I tried to use the author’s more objective definitions.
I had to do a little sleuthing to make sure that I was interpreting each statement correctly. From what I can tell, #7 would be true, since Obama pushed through his health care plan, the stimulus, and something else that I cannot remember right now. Other presidents got a “true” here for doing less.
#8 requires more social unrest (I think like in 1968).
#11 requires a war victory. The capture of Saddam pretty much ruled Howard Dean out in 2004, so I wonder if bin Laden will become important before the election. I was a bit conservative there, mainly because the intel was obtained during Bush’s term.
But I could be wrong.
November 8th, 2011 at 10:36 pm
Fair enough, I didn’t read the original article.
November 9th, 2011 at 5:19 pm
Will Solyndra qualify as a major scandal? Does Occupy (fill in the blank) count as major social unrest? Will the incredibly charismatic Ron Paul suddenly surge to the front? (Seems as though every other GOP candidate has.)
November 9th, 2011 at 5:26 pm
Laura, your point about Nate Silver in general is fair (though he’s honed his methods through the ’08 election, where they performed well), but the main point about the articles I linked is that they directly engage Lichtman’s methods and conclusions. Some of his critiques seem pointed to me. Lichtman gets his chance to respond in the second link.
November 10th, 2011 at 2:42 pm
Matthew, I have now read Nate Silver’s posts in more detail and agree that they are really nice. Silver brings up some valid points, and I don’t seriously defend Lichtman’s model. The one thing that Lichtman’s 13 Keys does very well is predict one year in advance. So if I want to blog about who will win in 2012, I have a model (albeit an imperfect one) to do so. I plan to write more about election forecasts in the next year.
Paul, I will have to redo the 13 Keys analysis after the GOP nominates their President to account for charisma and potential social unrest (it’s possible that there will be enough to turn that True into a False). I do not expect a scandal like the shenanigans that happened during the Clinton administration.
Things always come up that will change a few votes. The question is will they be enough to change who we will elect? The 13 Keys offers some insight, but I hardly think that it is the final word. The discussion here suggests that some of the Keys could have been interpreted either way, leading to another close election. It should be a lot of fun this coming year.
November 11th, 2011 at 1:10 pm
Laura, as your other post on Nate Silver isn’t up yet, I’ll add some links here that you might find interesting.
Silver’s NYTimes article describing his model: http://www.nytimes.com/2011/11/06/magazine/nate-silver-handicaps-2012-election.html?_r=1&pagewanted=all
Brendan Nyhan and Jacob Montgomery comment: http://www.brendan-nyhan.com/blog/2011/11/a-comparison-of-presidential-forecasting-models.html
Andrew Gelman comments: http://themonkeycage.org/blog/2011/11/10/forecasting-2012-how-much-does-ideology-matter/
Links courtesy of Ezra Klein: http://feeds.washingtonpost.com/click.phdo?i=5ae13b26b83c35ac5d93c85a128bffc9
November 11th, 2011 at 4:16 pm
And here’s another interesting, relevant link about yet another model: http://www.realclearpolitics.com/articles/2011/11/11/the_fuzzy_math_and_logic_of__election_prediction_models_112042.html
December 4th, 2011 at 12:02 am
obama will be reelected by the media and the sheeple will follow as usual.
January 9th, 2012 at 7:09 am
I honestly don’t see how Sarah Palin could be mentioned as a candidate to beat Obama. Anyone who would actually vote for Sarah Palin would vote for whomever is running against President Obama. She has zero ability to steal votes away from Obama. She may be charismatic to some “News” channels, but politically she is a joke, and would lose by a landslide.
If Mit Romney gets the nomination, I think he gets in a tight race with Obama, but I think he still loses.
Ron Paul may actually have the best chance to win the general election since he will still get the votes needed from republicans since their #1 goal is to make Obama a 1-term president. But on top of that, he has great charisma (perhaps not in speech and presence, but in the content of his message) with the younger generations which typically would be expected to favor Obama. He steals more votes from Obama than any other republican nominee.
I honestly think Ron Paul wins decisively, Mit Romney loses in a tight race, and all others (including hypothetical Sarah Palin runs) lose undoubtedly.
January 9th, 2012 at 8:58 am
@PiP: My comment about Sarah Palin was meant to be a bit tongue in cheek. I meant to be a little controversial, since I wrote that post a year before the next Presidential election. I was surprised that–according to this model at least–that a Sarah Palin candidate would have a good shot. But that could be a symptom that something is wrong with the model rather than a reflection of reality. I’ll have more grounded posts on politics coming up.