About this Blog
About The Author
Email Me

RealClearPolitics HorseRaceBlog

By Jay Cost

« Swing State Review: Virginia | HorseRaceBlog Home Page | Hats off to the Obama Campaign »

Thoughts on the State of the Race

With Barack Obama about to make his vice-presidential selection, the general election is finally set to begin. Thus ends the middle period between the nomination campaign and the general election campaign. I would like to offer some reflections on the state of the race, the media, the candidates, etc.

Once again, the mainstream media disappoints.

When I started my blog in 2004, I did so because I was sick and tired of the way the mainstream media covered the presidential campaign that year. Unfortunately, but not unsurprisingly, they are committing the same errors this cycle. I expect them to continue to.

I could catalog endlessly all of the things I think the media mishandles, but ultimately it comes down to how they mischaracterize the average voter, who appears to be like somebody you'd hear about in a Kinks tune - a sillyhearted schizophrenic with amnesia.

Take an example, one making its way through the mainstream media at the moment, this McCain housing "gaffe." The only way this non-event could have any significant effect on the election is if it influences the vote choices of a good number of citizens. But think of what we must assume about those citizens to believe that! What kind of a moron would somebody have to be to be planning to vote for McCain, hear that he doesn't know how many houses he has, and presto! that voter switches to Obama. Of course, his memory is so shoddy that he can't keep the decision in his pea-sized brain through the next mini-scandal. Sooner or later, Obama will do something stoopid - and presto! he's back voting for McCain.

I'm not accusing anybody in the mainstream media of being intentionally condescending to the average voter. I think that most analysts just get caught up in their culture. They're not "plugged in" to the way average people think. They're plugged in to the way Washingtonian politico types think, which is very different.

I just don't think mainstream media people understand how average voters think about politics. They know that average voters pay relatively little attention. But after that they don't know very much. So, they develop an image of them that they don't realize is actually a caricature. And because Washington is the center of their universe, they're never forced to reconcile the caricature with the genuine article. Their only sustained interaction with average voters is through the public opinion poll. When you think about it, those polls are nothing more than aggregations of "1's" and "0's" given in response to a handful of pre-determined, hyper-structured questions. How much can you know if that's your only source of information?

But haven't the polls "tightened?"

First of all, let me say that I frickin' hate all these polls. Hate them. Public opinion polling is a great boon in many respects - but it has been grossly overused in academic and popular political analysis, to the detriment of both. It has become a substitute for a lively imagination and rigorous reasoning. The number of polls in this election has reached an absurd level, and it has had the effect of further impeding good political analysis, which was already in short supply.

As for the tightening, it depends on the poll you look at. Gallup hasn't budged. And, when I look at polls, I go to Gallup first. It's the gold standard. And, God bless them, they interview 2,500 people every three days. That means that the margin of error on their tracking poll is so low that statistical inefficiency is not a problem. The only problem could possibly be statistical bias. And if there is one pollster that is unbiased (in a strictly statistical sense of the term), I'm betting it's Gallup.

[Update, 12:30 PM: Speaking precisely, I should say that Gallup hasn't budged substantially. To quote the standard-bearers of Frank Gallup: "[Obama's one-point advantage] matches the average gap between candidates over the past week...Obama had enjoyed a slightly larger three-point average margin over McCain from the time Obama clinched the Democratic nomination in early June through the Aug. 11-13 tracking results." So, we've gone from Obama +3 to Obama +1. Somebody catch me 'cause I'm gonna faint! The point is that the race was close in June, close in July, and close in August. I see little difference between 5-, 3-, and 1-point spreads, at least in the summer.]

More generally, the fact that the polls occasionally respond to the MSM's discussion-of-the-day is largely irrelevant to my point. Poll responses and vote choices are two very different things. The mass public picks up the MSM's dialogue in some sense, regurgitates it to the pollster, and the MSM declares that the public does in fact respond to what the MSM thought it did. It's called the echo chamber.

Here's the state of the race, at least as I see it.

(1) The macro conditions favor the Democrats in a way we have not seen in at least 28 years.

(2) In response, the Democrats nominated a candidate with relatively little governing experience and a background quite different from white voters, who swing presidential elections.

(3) The Republicans nominated a candidate who built a national reputation by disagreeing with George W. Bush in particular and the Republican Party in general, in the hopes that this man is immune from the public disaffection with the GOP.

(4) The public now gets to choose a man with little experience and a different background, or a semi-Republican. They're not sure which one they want. And because there are two wars on, a credit crisis, a weak economy, and high gas prices - they're taking their sweet time in deciding.

(5) Anybody who tells you what is going to happen is probably trying to sell you something.

Did I miss anything?

History is of relatively little value in determining where this race is headed.

We can build a model that predicts presidential vote outcomes based on macro conditions. We can profitably take that back to 1948 or thereabouts. That gives us fifteen previous elections to work with.

But this is an open presidential election, one where the big dog is not running for reelection. Those are very different, and there have only been five of them since 1948.

In those elections, you'll usually see the vice-president running on behalf of the incumbent party. There's been just one exception.

That year was 1952. Structurally speaking, this year has a lot in common with 1952.

But the candidates have nothing in common with 1952. Instead, they are much more like the candidates from 1976. Barack Obama reminds me of Jimmy Carter - he's relatively inexperienced and his background is such that a segment of this country is probably going to balk at voting for him. John McCain reminds me of Gerald Ford, though I suspect he would have let Nixon go to jail.

Unfortunately, we've never had a previous presidential election where the structure is 1952 and the candidates are 1976.

Bottom line: we're in unchartered water here. History is still useful, and it establishes that the Democrats are favored. But the limitation of history is that we don't know how heavily they are favored.

-Jay Cost