In 2005, FairVote issued its first comprehensive report on presidential elections using our measure of state partisanship, The Shrinking Battleground, which in 2006 was published in tandem with other reports in Presidential Election Inequality.

This analysis provides the foundation for our critique of our current Electoral College system. It makes a convincing case that the number of "swing states" is declining and their definition is becoming more rigid - a pattern that has only been reinforced by elections in 2008 and this year.

We have made use of a key element of this analysis to help those interested in Tuesday's election to predict its ultimate outcome using the vote-margin in each state on election night. We show each state's Republican partisanship for 2004 and 2008 - that is, a projection of the vote share the Republican nominee for president would have won that year if the national popular vote had been an exact tie.

We have set up a blank space on the right of the chart below where you can write the percentage by which news sources are reporting the candidates are winning in each state. Compare this margin with the "2012 benchmark" provided in the chart below, which should be used to give you an idea of whether a candidate is doing better in a state than expected. The more that one candidate is surpassing their benchmark, the more likely it is that this candidate will win the national popular vote and win the election.

Of course, there are no guarantees that the national popular vote winner will in fact win the election. For that determination, we have bolded the eleven most important swing states in this election, since the candidates' performance in these particular states will be the most telling for who will win a majority in the Electoral College.

In 2005, FairVote issued its first comprehensive report on presidential elections using our measure of state partisanship, The Shrinking Battleground, which in 2006 was published in tandem with other reports in Presidential Election Inequality.

This analysis provides the foundation for our critique of our current Electoral College system. It makes a convincing case that the number of "swing states" is declining and their definition is becoming more rigid - a pattern that has only been reinforced by elections in 2008 and this year.

We have made use of a key element of this analysis to help those interested in Tuesday's election to predict its ultimate outcome using the vote-margin in each state on election night. We show each state's Republican partisanship for 2004 and 2008 - that is, a projection of the vote share the Republican nominee for president would have won that year if the national popular vote had been an exact tie.

We have set up a blank space on the right of the chart below where you can write the percentage by which news sources are reporting the candidates are winning in each state. Compare this margin with the "2012 benchmark" provided in the chart below, which should be used to give you an idea of whether a candidate is doing better in a state than expected. The more that one candidate is surpassing their benchmark, the more likely it is that this candidate will win the national popular vote and win the election.

Of course, there are no guarantees that the national popular vote winner will in fact win the election. For that determination, we have bolded the eleven most important swing states in this election, since the candidates' performance in these particular states will be the most telling for who will win a majority in the Electoral College.