The October 2008 edition of the journal PS: Political Science and Politics included a symposium entitled “Forecasting the 2008 National Elections.” This symposium included a set of articles in which more than a dozen political scientists predicted the outcome of the presidential election (who would win and by what percentage of the popular vote). Though the articles were published in October, because of the time necessary to review and edit the manuscripts and prepare them for publication, they were all completed by the middle of the summer. In fact, all of the data on which the predictions were based was available prior to Labor Day, the traditional beginning of the general election campaigns. This is important because it indicates that all of the predictions in each of the articles preceded the beginning of the general election, so neither the general election campaign of then-Senator Obama nor Senator McCain influenced the results or predictions in any way.

As each article included its own particular model for predicting the partisan distribution of the popular vote, it should come as no surprise that there was some variation in the predictions. While some symposium participants thought the election would result in a relatively easy victory for Obama (see Abramowitz 2008, Holbrook 2008), others thought Obama would win in a nailbiter (see Cuzan and Bundrick 2008; Norpoth 2008). The editor of the symposium, James Campbell, actually predicted a very close McCain victory.