As a result, pundits and prognosticators say that there are only nine states that really matter in this year's presidential election: the so-called "battleground" states where the outcome is still uncertain enough to warrant attention from the candidates. As the Associated Press pointed out, modern campaigns now have the data to target voters even more narrowly than that, and they're now focusing on just 106 "swing counties" (out of 3,143 in the United States).
The reason, of course, is the winner-take-all system of the electoral college, which dictates that whoever wins a majority of the votes in a state gets all of that state's electors. In fact, the winner-take-all (or first-past-the-post) principle pervades American politics. As political scientists know, these sorts of electoral mechanisms tend to foster the creation of two-party systems. (The framers of the American constitutional system actually didn't want to have political parties at all, of course; but this is just another one of those cases where their design had unintended effects.)
The problem is that a two-party system doesn't come anywhere close to exhausting the range of options for political expression. Earlier this year, when pollsters decided to examine the motives of non-voters, one of the questions involved alternate political parties. Only 32 percent of non-voters agreed with the premise that two parties are good. Twenty-six percent of them said that a third party is necessary, while another 27 percent preferred "multiple" parties.
That's why it's wrong to dismiss non-voters as ignorant couch potatoes. Under the American system, a vote cast for a third party (the Libertarians, say, or the Greens) is a lost vote. Your ballot has no effect whatsoever on the actual balance of power, so abstaining from an election that offers any chance to pick the policies you'd like to see makes perfect sense. This is also why it's somewhat nonsensical to ask voters whether they'd vote for third parties under the current circumstances. What good would that do?