I'm one of those people who bangs his head against the wall at the press' extreme preference for horse-race political coverage rather than stories about the issues, so I feel extremely dumb for not having thought of this explanation before Andrew Gellman did: "My theory, at least for the general election, is that most of the voters have already decided who they're going to vote for--and even the ones who haven't decided are often more predictable than they realize." And of course on tap of that, the truest "swing" voters are precisely the ones least likely to be paying attention to political coverage in the media.
« Pronounciations | Main | Why Am I Not Surprised? »
Why Horse Race
05 Nov 2007 04:02 pm
Comments (24)
Two more thing: Journalism often construes its job as "reducing" uncertainty, and nothing reduces uncertainty more than figuring out who's winning.
Second, there's the phenomenon that the stump speeches change very little, and when they do change they change very slowly, and so on-the-trail journalists have little to report on. This leaves the horse race as one of the few things to cover.
Third there's the lack of expertise phenomenon: most journalists don't know much about economics or health policy or foreign policy, and it's hard to report on these fields without knowing who's real and who's feeding you a line.
In theory a campaign might be able to change this dynamic by using surrogates effectively, but it would be very hard and no one seems interested in doing so.
Matt,
Analogously, why are you so interested in the horse race aspects of the NBA?
Because, like most human beings, you are highly interested in who wins. It's an inherently absorbing topic. As Lenin said, the ultimate question is always "Who? Whom?"
Personally, I don't care who wins in the NBA. To the extent I'm interested in the NBA, I'm only interested in "the issues," such as what type of strategy works better, what the influence of hip-hop has on the style of play, what we can learn about human biodiversity from this collection of super-tall individuals, what we can learn about race relations from the NBA, and so forth.
But, that's a minority point of view, to say the least!
My theory is that Andrew Gellman has never worked in a campaign. People do change their minds. Things change a lot over the course of several months.
He's more or less suggesting that most people who claim to be open to having their minds changed--or who claim to have changed their minds--are lying or self-deluded. He's probably mostly right.
In the case of the GOP primary race, however, I think the sports analogy breaks down.
Is it objectively true that most Republican voters are currently locked and loaded on their single favorite candidate and thus are only interested in the primary race for the same reason that sports fans are interested in following their team? I sure don't get that impression.
Gellman proposes an explanation that only applies to the general election. If horse race reporting dominates just as much in the primary phase (which I'm sure it does) then his explanation is worth nothing (which I'm sure it is).
The real explanation, of course, is what Steve Sailor said above. Can anyone out there think of a news story more boring than one analyzing the differences between the Clinton and Edwards health care proposals?
Someone with time on their hands could do an easy experiment: compare the number of stories about a campaign with the projected closeness of the campaign. For example, were there a lot more presidential campaign stories in 2000 and 2004 than in 1996 and 1984? If so, then I think the sports analogy works pretty well.
A lot of what people are interested in is displays of personal dominance, which is what these debates are mostly interesting for: Can you come up with a quick comeback to put some challenger in his place? Are you The Man?
This is hardly an unreasonable topic of interest. What's the point of electing somebody President who gets pushed around by the people he meets? Then you aren't going to get anything like the policy issues you elected him for.
An example might be Ramsay MacDonald's last term as Prime Minister during the early Depression, when he was going senile and he was totally pushed around by the Tories. A Labour voter would be right to regret having elected a man unable to stand up for himself.
Dude, we've been over this. Well, Ezra Klein and James Fallows have:
http://ezraklein.typepad.com/blog/2007/07/what-james-fall.html
When ordinary voters ask candidates questions, they ask about real issues and hardly ever the horse race.
What southpaw said.
Also, there are many areas where people's choices are based on assumptions about the candidates, which might or might not be true. I think there's a natural tendency of people to assume that if the MSM isn't saying anything to the contrary, then Major Candidate X's approach to Major Issue Y is at least semi-responsible.
So you have an electorate full of people who have no awareness of the yawning chasm between the Democrats' plans to deal with health care and climate change, and the Republican candidates' complete absence of any such plans, because the MSM hasn't said shit about it.
If they actually did this basic reporting, you bet it would change some minds.
It has nothing to do with whether voters have made up their minds and everything to do with ratings.
Viewers hear "health care," "social security" and "Iraq" and change the channel or fall asleep.
They hear "Obama called Hillary stupid" and they might stick around a few extra minutes.
well, i think Jim W at 4:45 gets at the topic here, which is why this is a sensible explanation for press behavior.
you see, i can imagine something massively more boring than an article dedicated to telling me about differences in approaches to health care; for example, yet another turgid piece of horse-race reporting.
but over there on the publisher's side of the business, they are convinced that there are far more eyeballs who will look at an article about the horse race (even though i find it boring) than an article about health care proposals (that jim w finds boring).
it's not clear why they think this, by the way: how would they know?
in an attempt to accentuate the positive, though, i'll note, as someone up above did, that this love for horse-race reporting has opened up a new niche - informed commentary - that the blogosphere is filling. the opportunity to learn about the differences between clinton's and edward's health-care proposals is so much more available today than it would have been even 10 years ago.
Also, covering the horse race rather than the issues is an easy way for journalists to feel "neutral," "balanced," and "above the fray," which seems to be what they think people want and respect.
Two wrinkles to this:
1) Horse race politics can be covered in a less overtly "partial" way; whether that actually holds up is debatable, but if you're critiquing issues and policy, you're showing a certain ideological favoritism (that this is taken on faith in the media makes me, too, want to beat my head against a wall).
2) Much bigger issue to me.
I suspect this perception that Gellman is offering has some truth to it, but only to a point. It's probably very strong in terms of voter preference today vs. tomorrow. Or tomorrow vs. next week. Or maybe even next week vs. next month. But this is a real forest & trees problem.
One of the interesting arguments Page & Shapiro make about public opinion is that it is, generally, rational; people's opinions vary high or low around weakly held central tendencies. But because there is a small amount of variation, those opinions can migrate. So while someone might be weakly centered dislike of a policy, varying between strong dislike and slight favor, he or she could theoretically be moved to weakly centering around indifference, varying from average disfavor to average favor. And people could do this without ever actually feeling like they change their minds, because at any given point in time, the attitudes they held yesterday and the attitudes they hold today are within the same comfortable, variable range those attitudes have been in for a couple months.
The problem here is oddly similar to the climate change problem. Obviously, if people could feel catastrophic effects coming, or if they could really feel the earth getting notably warmer, they'd be willing to make the necessary sacrifices because they would see it in their own interest. But the changes are so small and incremental that it's hard to perceive them. Similarly, I suspect the news media would be happy to make big policy pushes if they could see it making a difference. But they can't see it! They look at candidate polling in monthly terms, not semi-annual terms. They look at candidate polling vis-à-vis their coverage in even shorter terms, because they don't have longer, concentrated, policy-based discussions. So they can only pick very minute data points to compare.
Well of course they can't see a difference! People don't want to think of themselves as too easily malleable. Our opinions are changeable, but only insofar as we can rationalize our changes. For most of us, either a cataclysmic event (9/11) or a very long period of change is required. So we resist changing in the short term in any way large enough to actually feel like we're changing.
He's more or less suggesting that most people who claim to be open to having their minds changed--or who claim to have changed their minds--are lying or self-deluded.
Even if this were true, polls certainly don't measure it. Otherwise we could take one poll at any time of your choosing and go home.
We wouldn't see swings in popularity over time, we wouldn't speculate about the coming powerhouse of Newt Gingrich or Fred Thompson or Al Gore. We'd be sitting here debating the reasons Joe Lieberman lost the 2004 election to George Bush. Or discussing George Sr.'s crushing defeat of Paul Tsongas 12 years earlier.
Since polls are as uncertain as the delusions of the population being questioned the constant obsession with the horserace is still content free. So you may as well just focus on the issues.
As to your original question (why does the media focus on the Horse Race), the answer is really quite simple.
Because they're lazy.
If they actually did this basic reporting, you bet it would change some minds.
Fair enough.
Since polls are as uncertain as the delusions of the population being questioned the constant obsession with the horserace is still content free. So you may as well just focus on the issues.
How did I wind up defending the news media's retarded coverage? Yikes. I may be a cynic about people in general, but I truly didn't mean to suggest that that excuses our terrible news coverage.
Of course Jim W is also right about primaries.
It's impossible to justify the way the media covers elections on the basis of voter preferences. That's the same logic behind the claims that there's a bunch of crap on the TV because that's what the viewers want. Right. The way the media covers politics has nothing to do with voters and everything to do with filling up the paper (or the news hour) without expending too many resources and without overtly taking sides -- though, of course, the media does take sides (they want us all to know that the Republicans are mas macho). Also, Maureen Dowd is far from the only so-called journalist who prefers gossip to welfare reform.
The voting behavior Gelman’s referring to is very well-established, so anyone working themselves into a huff about whether ‘he’s worked on a campaign’ or whatever needs to get past personal anecdotes and look at actual research in the field.
That said, I don’t quite buy his argument. It seems like he’s making a sort of Revealed Preference argument, that what’s available must represent underlying demand. I’m not sure how directly Big Media political coverage is exposed to market forces, though. As much as we obsess over it, the political beat is a tiny part of the overall supply of news about car accidents, celebrities, weather, etc. And the single-subject political shows that do live and die on ratings rarely feature much of the horserace coverage that drives us all crazy.
I’m personally more inclined towards an institutional argument. Given the central role of objectivity, “balance”, etc. in the press’s self-conception, serious policy reporting is risky, since it’s almost impossible to avoid the appearance of taking sides.
Semi-OT, I’ve been going through the new(ish) edition of Gelman’s Bayesian Data Analysis, and it’s aces.
I agree with Jinchi, a rarity, that Gellman is wrong: Even if this were true, polls certainly don't measure it. Otherwise we could take one poll at any time of your choosing and go home.
We wouldn't see swings in popularity over time, we wouldn't speculate about the coming powerhouse of Newt Gingrich or Fred Thompson or Al Gore. We'd be sitting here debating the reasons Joe Lieberman lost the 2004 election to George Bush. Or discussing George Sr.'s crushing defeat of Paul Tsongas 12 years earlier.
Gellman completely ignores major intraparty swings are the norm.
Gellman ignores big shifts in the Presidential election itself.
And these shifts are Pol SCI 101. Why did Truman overcome a 20 point defecit and beat Dewey. Why did Bush I overcome and 18 point deficit to Dukakis. How did Reagan get a landslide from what started as a slight lead over Mondale following the 2004 Dem Convention?
Why do little known or "just strong in a single region of America governors or VPs in the background of most voter's minds" usually end up President when Gellman claims "everyone knows who they want to vote for before Primary elections are even held". Mainly because the national polls track Q Factor - how well a candidate is known nationally before the race starts.
(Which is why you always see camera-loving Senators pop up every 4 years on seeing how they are Destined, yes Destined to be President based on Q Factor polls. And almost never are. 198 sitting members of Congress have run in the last 100 years. 2, JFK and Warren Harding, got in.)
That said, there is plenty to lambaste the press for in keeping US voters ill-informed.
1. Horserace and momentum are favored over issues. Whole races can go by with no coverage of great differences between the candidates on the things that most affect Americans outside the Media elites - education, health care, jobs.., cultural threats..
2. Media focus on "national numbers" misses critical state polls that end up. In 2000, the media worked how Gore had a 3 point lead outside any polling errors in the popular vote, had huge margins in the biggest state, and was ahead in the "critical swing states outside Ohio and possibly Florida".
They completely missed that Gores new anti-gun fervor led voters away from him in Arkansas, Tennessee, Florida, New Hampshire, and almost - New Mexico.
if it bleeds it leads... horse-race style reporting allows breathless coverage of debate gaffes and latest poll numbers. Covering issues and policy statements in the allocated 30-75 seconds on TV or 1200 word column is very difficult -- imagine discussing corporate welfare in 1200 words and capturing the details that matter. It requires research, multiple interviews, and a lot of BS testing by researchers -- too many news cycles for a very low payoff to the media.
Maybe that's why I subscribe to the atlantic? every once in a while, you get a multi-page! review of an important issue.
Re: "When ordinary voters ask candidates questions, they ask about real issues and hardly ever the horse race."
There are a couple logical fallacies packed into this. Here goes.
First of all, I doubt that the miniscule subset of voters who actually ask questions of candidates are an unbiased sample.
Secondly, what else are they going to ask of the candidates themselves? If I had an audience with the Pope, I'd probably ask him a question about religion even though I have almost zero interest in it. How about when ordinary voters ask questions of pollsters? Do they ask about real issues or the horse race?
A final point is that there is intrinsically more news in the horse race than in the issues. This is because the candidates positions on the issues remain mostly static over the course of a 1+ year campaign, whereas the horse race stuff keeps changing.
Secondly, what else are they going to ask of the candidates themselves?
Let's see, we could ask Brian Williams:
MR. WILLIAMS: Specifically, what are the issues where you, Senator Obama, and Senator Clinton have differed, where you think she has sounded or voted like a Republican?MR. WILLIAMS: Senator Obama, was Senator Clinton's answer to the opposition of the Iraq war question consistent, in your view?
MR. WILLIAMS: .. And we're going to start with another subject at the top of this segment. Senator Clinton, it will go to you. It speaks to electability
Or how about Chris Wallace:
WALLACE: Gentlemen, you all have a couple of things in common. You all seem to be planning to run against Hillary Clinton in the general election and at this point all of you -- I repeat all of you -- are losing to her in the polls. So let's talk about how you intend to beat her.WALLACE: Congressman Paul, you're against the Iraq war. So is Senator Clinton. So what are the differences between you?
WALLACE: Is Hillary Clinton fit to be commander in chief?
WALLACE: Mayor Giuliani, you trail Clinton by four points in our latest Fox News/Opinion Dynamics Poll.
WALLACE: Governor Romney says the Republicans aren't going to beat Hillary Clinton by acting like Hillary Clinton. And the point seems to be that on a lot of the social issues, like abortion and gay rights and gun control, that there's not much difference between you and Clinton. Is there?
WALLACE: Mayor, Mayor, I’m going to give you 30 more seconds because I want to – you made a big point on the campaign trail about how you know how to beat Senator Clinton. But the fact is, I looked today at the latest polls in all the key battleground states: Florida, Pennsylvania, Ohio – you trail her in all three.
Comments closed November 19, 2007.
| 
Heads in the Sand
My theory is that Andrew Gellman has never worked in a campaign. People do change their minds. Things change a lot over the course of several months.
Policies have consequences. There are winners and losers. Understanding them, never mind explaining them, requires a great deal of work. Moreover, the corporate management at most news organizations are not eager for ordinary readers to understand the trades offs of, say, chosing the Iraq war over single payer healthcare. That is why it is endless horse race commentary.
That is the vacuum that created Joshua Micah Marshall's opportunity.
Posted by Alice | November 5, 2007 4:15 PM