Who’s going to win?

by Russ Roberts on November 4, 2012

in Politics

The state polls say Obama though there appears to be some tightening.

The national polls say Romney, though they are all pretty close.

So one answer is that Romney will win the popular vote and lose the electoral college.

For a fantastic analysis of polling generally and the precision or lack thereof, of polls, and the Nate Silver excitement, see this post by Dan McLaughlin. An excerpt:

Polling is “scientific,” in the sense that it attempts to follow well-established mathematical concepts of random sampling, but political polls remain as much art as science, and each polling cycle presents different challenges to pollsters’ ability to accurately capture public sentiment. Quick summary: dating back roughly to George Gallup’s introduction of modern political polling in the 1936 election, a pollster seeks to extrapolate the voting behavior of many millions of people (130 million people voted in the 2008 presidential election) from a poll of several hundred or a few thousand people. In a poll that seeks only the opinion of the public at large, the pollster will seek to use a variety of sampling techniques to ensure that the population called actually matches the population as a whole in terms of age, gender, race, geography and other demographic factors. In some cases, where the raw data doesn’t provide a random sample, the pollster may re-weight the sample to reflect a fair cross-section.

Political polling is a somewhat different animal, however: not all adults are registered voters, and not all registered voters show up to vote every time there’s an election. So, a pollster has to use a variety of different methods – in particular, a “likely voter” screen designed to tease out the poll respondent’s likelihood of voting – to try to figure out whether the pollster’s results have sampled a group of people who correspond to the actual electorate for a given election. This is complicated by the fact that voter turnout isn’t uniform: in some years and some states Republican enthusiasm is higher than others, in some Democratic enthusiasm is higher than others. You can conduct the best poll in the world in terms of accurately ascertaining the views of a population that mirrors your sample – but if your sample doesn’t mirror that season’s electorate, your poll will mislead its readers in the same way that the Literary Digest’s unscientific poll did in 1936, or the RCP averages in the Senate elections in Colorado and Nevada in 2010, or the polls that failed to capture the GOP surge in 2002.

Read the whole thing. It’s fantastic. It parallels my view of econometric analysis as being rather imprecise.

My intuition is that Romney is going to win. I base that on two ideas. One is that the economy just isn’t doing very well. For that to be overcome by Obama, he would have to run an excellent campaign and Romney would have to run a bad campaign. Given the unemployment rate and the growth rate of the economy, Obama has run an excellent campaign painting Romney as a rich and dangerous man who cares little about the rest of us and who will return us to “the policies that got us into this mess.” (That phrase drives me nuts by the way–does he mean the Bush tax cuts? Obama extended them. The repeal of Glass-Steagall? That was signed by Bill Clinton with bipartisan support.) I’ve never thought Romney was a particularly good campaigner, so Obama’s attempts to paint him as a frightening alternative might have succeeded. That changed with the debates. Romney may not give a particularly thrilling stump speech but he obviously did very well in the debates. He came across as a reasonable guy rather than a scary guy.  So the economic fundamental have always favored Romney and Obama has not been able to sell the Romney as scary guy story after the debates.

The second point is that Romney is dominating the independent vote. If those polls are accurate and if Romney easily carries the independent vote as he appears to be doing now, then the only thing that can save Obama is a healthy turnout of Democrats relative to Republicans. But everything points to lethargy among the Democrats. They will of course vote for him, but they are holding their nose. And some of them will simply not vote. Turnout among young people will not be what it was in 2008. Turnout among African-Americans will not be what is was in 2008–unemployment among African-Americans is over 14%. And as a ridiculously anecdotal piece of evidence, my neighborhood which is in Montgomery County, Maryland–as blue a place as you can imagine outside of say Cambridge Massachusetts or Berkeley California–has no Obama yard signs at least on my usual walks and drives through the neighborhood. I know, it’s a silly data point but it’s a bit weird. There are three Romney yard signs on my usual path. That’s even weirder than no Obama signs. The bottom line is that enthusiasm for Obama is very low.

As a commenter, Sam, here at the Cafe once said (I’m doing this from memory), there are a lot of dots in the universe. You can connect them any way you want to tell a story. The question isn’t whether the dots you’ve connected are really there, it’s why have you left out the other dots? (And Sam, I’d love to know your full name so I can attribute that quote more precisely). So what dots have I left out?

I think there are two important dots. The state polls and the betting market. Intrade right now, noon on the Friday before the election, has Obama winning easily–he has a 67% chance of victory. The wisdom of crowds says Obama.

And what about those state polls? Polling isn’t a science but you can’t just dismiss them. See the Dan McGlaughlin post for some interesting reasons why they might be wrong. I have a different idea. It comes from Doug Rivers. Pollsters don’t like being very different from their competitors so they often fudge the turnout numbers so they don’t go out on a limb. Going out on a limb is great if you’re right. If you’re wrong, you look like an idiot. If you want to have an ongoing business, you don’t want to go out on a limb too often. So what looks like a consensus–an average of independent results, can actually represent, groupthink But that’s true of the national results, too.

Intuition is dangerous, especially mine. Or yours. Sometimes intuition is just another name for cherry-picking. So going against my intuition is this result from YouGov (where Doug Rivers is the Chief Innovation Officer). I think YouGov is the best polling group for dealing with the no-one-uses-their-landline problem and the challenges of actually figuring out who’s going to vote. YouGov still has Obama in the lead and suggests that the debate bounce for Romney wasn’t real. The post is from October 23 but that’s late enough to be a bad sign for Romney.

Intuition is a fancy name for guessing. Guessing is cheap. It’s an investment in being able to tell people Wednesday morning how smart you are.

The only thing I really know is that if the Republicans can’t beat an incumbent with lousy economy numbers, who has a disillusioned base and who has run the national debt to $16 trillion in a country that trends conservative, they have some soul-searching to do on who might do better next time.

Be Sociable, Share!



Add a Comment    Share Share    Print    Email

Previous post:

Next post: