A Non-random Walk Down Campaign Street

by Andrew Gelman on September 28, 2012 · 6 comments

in Campaigns and elections

Political campaigns are commonly understood as random walks, during which, at any point in time, the level of support for any party or candidate is equally likely to go up or down. Each shift in the polls is then interpreted as the result of some combination of news and campaign strategies.

A completely different story of campaigns is the mean reversion model in which the elections are determined by fundamental factors of the economy and partisanship; the role of the campaign is to give voters a chance to reach their predetermined positions.

The popularity of the random walk model for polls may be partially explained via analogy to the widespread idea that stock prices reflect all available information, as popularized in Burton Malkiel’s book, A Random Walk Down Wall Street. Once the idea has sunk in that short-term changes in the stock market are inherently unpredictable, it is natural for journalists to think the same of polls. For example, political analyst Nate Silver wrote in 2010:

In races with lots of polling, instead, the most robust assumption is usually that polling is essentially a random walk, i.e., that the polls are about equally likely to move toward one or another candidate, regardless of which way they have moved in the past.

But, as I discussed then in the context of remarks by Silver and Justin Wolfers, polls are not stock markets: for many races, a forecast from the fundamentals gives a pretty good idea about where the polls are going to end up. For example, in the 1988 presidential election campaign, even when Michael Dukakis was up 10 points in the polls, informed experts were pretty sure that George Bush was going to win. Congressional races can have predictable trends too. Political scientists Erikson, Bafumi, and Wlezien have found predictable changes in the generic opinion polls in the year leading up to an election, with different patterns in presidential years and off years. Individual polls are noisy, though, and predictability will generally only be detectable with a long enough series.

Noah Kaplan, David Park, and I have a paper on the topic (to appear in Presidential Studies Quarterly) reviewing the literature and analyzing data from polls during the 2000, 2004, and 2008 elections. We show that, as the campaign progresses, vote preferences become more predictable based on fundamental variables such as political ideology and party identification. This is consistent with a “mean reversion” model in which the campaign serves to solidify latent preferences, but it is not consistent with a random walk model in which a campaign is an accretion of unpredictable shocks.

To many of the readers of this blog, the above is not news. Political scientists have been talking about “the fundamentals” for awhile, to the extent that journalists and other observers sometimes overestimate the importance of the economy in determining the election (for example, here’s a clueless history professor likening the predictability elections to “the law of gravity”). As John Sides explained reasonably, you have to be careful when translating economic numbers into vote predictions.

Still, a bit of old-fashioned random-walk thinking remains in the old-fashioned news media. For example, Michael Kinsley recently wrote:

In 1988, Michael Dukakis, who was ahead in the polls just after the Democratic convention, declared in his acceptance speech: “This election isn’t about ideology. It’s about competence.” Then he proceeded to blow a large lead and lose to George Bush the Elder, who turned out to be a tougher old bird than anyone suspected.

This sort of understanding of campaigns was pretty standard a few decades ago, back when Kinsley was editor of the New Republic, but nowadays we wouldn’t frame Dukakis as having “blown a large lead” but rather that he lost a lead that was effectively unsustainable, given the economic and political conditions of 1988. Nor would we need to characterize Bush Senior as a “tough old bird” for winning this election; it was more about being in the right place at the right time.

To say that Dukakis blew a lead is not quite to buy into a random-walk model, but I think it is close. Given what we know about elections, I think it would be more accurate to say that the 1988 election was Bush’s to lose (and he didn’t).

Anyway, that Kinsley quote is an example of why I think this blog post could be helpful. I’m hoping that, by explicitly stating the random-walk and mean-reversion scenarios, I can make people more aware of the implicit models that underly their stories about campaigns and elections.


Brian Silver September 28, 2012 at 12:20 pm

Andrew, haven’t read your paper yet but you write here: “We show that, as the campaign progresses, vote preferences become more predictable based on fundamental variables such as political ideology and party identification.”

I have a question that relates to whether polls should weight results by party ID — an issue that, as you know, has inspired the RW to attack “liberal” pollsters because they have too many Democrats in their samples (when practically no pollster — Rasmussen is a prominent exceptioin — actually weights their results by party ID).

Namely, if party ID is itself affected by events, and it changes over time, what are you assuming about its stability and its priority as a predictor of the vote, rather than just a correlate?

Scott Monje September 28, 2012 at 2:06 pm

“. . . inspired the RW to attack “liberal” pollsters because they have too many Democrats in their samples . . .”

Curiously, Fox News seems to be part of the conspiracy. Their poll shows Obama leading 48 to 43 among likely voters, 49 to 41 among registered voters. (Their write-up says Obama’s lead is within the margin of error, which the report itself puts at plus or minus 3 percentage points.)


Andrew Gelman September 28, 2012 at 3:14 pm


I do think it makes sense to weight by party ID. Party ID changes over time, but that can be modeled. Cavan Reilly and I published a paper on this back in 2001, with the catchy title of “Post-stratification without population level information on the post-stratifying variable, with application to political polling.”

Nadia Hassan September 28, 2012 at 5:03 pm

For what it’s worth, Professor Gelman, Bill Galston and Gary Langer talked about the debates and said this:

“With the debates approaching – albeit their importance debatable – the most perplexing element of the 2012 presidential campaign is this: Barack Obama could win.
The condition of the economy suggests otherwise. The general rule is straightforward; when economies are bad, incumbent presidents pay the price. This year, it’s not so simple.
The absence of a compelling economic plan from Mitt Romney, doubts about his priorities, his shortfall in personal popularity and his campaign missteps have combined to make the contest a more competitive one than 8.1 percent unemployment would predict.”

Bill Galston: “Rather than sustaining a consistent theme, he lurched from issue to issue in response to the events of the day. He failed to counter the entirely predictable attacks on his leadership at Bain. He allowed the controversy of his tax returns to linger. He compounded these felonies with a seemingly endless series of gaffes, capped by a pratfall-filled foreign trip. His convention was by a considerable margin the least effective in decades. His selection of Paul Ryan shifted attention from his greatest potential strength—the economy—to the House Republicans’ politically toxic budget. And for a man whose supposed calling-card was managerial competence, Romney has run his own campaign very poorly indeed.”

Eli Rabett September 29, 2012 at 8:57 am

The unemployment rate in Fall 1984 was ~7.5% as opposed to 8.2% today. A difference, but not that huge a one.

Peter Principle September 29, 2012 at 12:50 pm

It’s interesting that the random walk hypothesis has opposite real world implications in the political sphere than it does in the financial markets.

In finance, the random walk implies that active management — stock picking — is useless, since any success by definition will be due to luck, not skills. Better to throw darts.

In politics, though, the random walk implies that races are won and lost based on what happens during the campaign — thus the need for an adroit campaign strategy to avoid “blowing” a lead or to make the other guy “blow” his.

The fundemantals-driven model, on the other hand, implies that markets are inefficient (if not, the underlying fundamentals would already be reflected in prices (poll results), producing yet another random walk.

Inefficient financial markets imply opportunities to earn alpha (hurray for hedge find managers,). But in politics, they imply that the fundamentals will out, no matter how good/bad a campaign the candidates run. At most, a good campaign can only “allow” the underlying fundamentals to do their thing.

In other words, a bad campaign can hurt you, but a “good” campaign can’t really help you.

I guess we need a Jack Bogle of politics to preach the virtues of campaign indexing.

Comments on this entry are closed.

Previous post:

Next post: