In the context of a fascinating, detailed account of the sex scandal that ended the presidential candidacy of Gary Hart, political journalist Matt Bai writes:
Back then, Hart was as close to a lock for the nomination — and likely the presidency — as any challenger of the modern era. According to Gallup, Hart had a double-digit lead over the rest of the potential Democratic field among Democrats and Democratic-leaning independents. In a preview of the general election against the presumed Republican nominee, Vice President George H. W. Bush, Hart was polling over 50 percent among registered voters and beating Bush by 13 points, with only 11 percent saying they were undecided. He would have been very hard to stop.
This is just wrong. Whoever won the Democratic nomination was highly unlikely to win the presidency, a topic that Gary King and I discuss in our epic-length 1993 paper, “Why Are American Presidential Election Campaign Polls So Variable When Votes Are So Predictable?” You know who else was leading Bush in the polls by 13 points? Michael Dukakis, and you know what happened to him. When the election finally came around, Bush beat Dukakis by nearly eight percentage points. Would a switch from wimpy Dukakis to tough-guy Hart have led 8 percent of Bush’s supporters to vote differently? I don’t think so, and I think it’s fair to say that my view represents the overwhelming opinion of political scientists, following decades of research by Bob Erikson, Steven Rosenstone, Doug Hibbs and many others on the predictability of presidential elections.
As Jonathan Chait (a political reporter of the next generation after Bai, and more influenced by political science) wrote:
Parties and candidates will kill themselves to move the needle a percentage point or two in a presidential race. And again, the fundamentals determine the bigger picture, but within that big picture political tactics and candidate quality still matters around the margins.
A percentage point, sure, but overcoming an eight-point margin in the general election? I don’t think so. As I wrote in my discussion with Chait, I’m pretty sure that Mike Dukakis, David Mamet, Bill Clinton and the ghost of Lee Atwater will disagree with me on this one, but Dukakis actually performed just as well in 1988 as predicted based on the fundamentals. There is no evidence he was out-campaigned by Bush, nor is there evidence that Hart was so awesome that he could’ve done so much better.
Let me emphasize that although general elections for president are largely predictable and the evidence is that they are decided by partisan and economic conditions much more than by candidates, the story is different for other sorts of elections. Here’s something I wrote a few years ago on why are primaries so hard to predict.
Political science and political reporting are complementary. The changing role of political media in the United States is an interesting and important topic, and it needs to be understood in the context of the immutable and anomalous details of particular examples, including the saga of Gary Hart. Hart’s story is dramatic, and Bai may be correct that it forever changed American politics. But it almost certainly had no effect on the outcome of the 1988 presidential election.