I have been reading up on US Presidential elections recently - or, to be accurate, I have been trying to find articles on recent elections, but have been disappointed with the poor quality of analysis done by scholars. What has happened, not only during the Bush Administration, but the Clinton Administration before it, is that researchers have discovered that "peer review" can be a very lenient thing if your peers share the same biases you carry yourself. And no, that's not to say that the researchers lean one direction politically, nearly so much as it seems they are always looking for hot new trends, and therefore little testing of assumptions is made. The 2004 election, therefore, has seen little serious examination, and given the present position I would argue that leaves room for a more open and productive discussion. I would be arrogant indeed if I proposed that I could write the definitive examination of that election, so I will simply attempt to add to the discussion, and for that I have simply been looking at the hardest of numbers, popular vote results. Strange as it may seem, even the most experienced of election analysts may forget that the election of the President of the United States is not one election, but is dependant on the results of 51 separate state and district elections. Consequently, while it may be more difficult to evaluate and even more so to explain, the separate results from each state and the District of Columbia must be evaluated in distinction.

One of the most obvious things from the 2004 election, was that both the Republican and Democrat candidates enjoyed more votes than their nominees enjoyed in 2000; turnout was obviously superior for the national totals. But a more interesting result occurs, when the percentage of the popular vote each candidate collected is considered for each state. Of the 51 contests, the Democrats' candidate improved his share of the popular vote in 21 states, maintained the same proportion in 11 others, and was lower in the remaining 19 states. The Republicans' nominee improved his share in 43 states, maintained the same proportion in 6 others, and was lower in just 2 states. This illustrates both the frustration in the Democrats' camp, as well as their myopia; both parties were able to improve their share of the popular vote in 2004 from 2000 in 13 states, because the increased focus on partisanship reduced interest in minor-party candidates. Therefore, improvement in the share of the popular vote in a state did not necessarily mean winning more states. The Democrats failed to understand this at the time of the election.

The results of the 2006 mid-term elections show the same effect in another light; the increased partisanship meant that if one major party suffered a loss of confidence in voter support, the other major party would benefit by that margin, which is what happened. The Democrats remained focused, and voters who supported Democrats in 2004 did so again in 2006. However, demoralized Republicans failed to support GOP incumbents to the degree they did in 2004, and this was the decisive margin in many contests. The demonstrated lesson of heightened party polarity is that driving voters to the major parties by reason of a "crucial" decision, requires the same or greater effort to get out the vote in subsequent elections, and at some point the effort must invariably fail, generally at the cost of the party in control. Whether this effect is only general or will also apply to the White House contest in 2008, must be considered according to the needs of that event. The purpose of this article is to show that this effect exists; subsequent analysis may or may not display evidence for trends or prediction.