That is, if the election gets held this week. Survey USA interviewed 600 registered voters in each of the 50 states, and it has published the Electoral College vote based on the results. Hillary Clinton edges John McCain by a mere 14 votes, while Obama tops him by 22. However, Survey USA has some caveats:

There are specific limitations to this exercise. The winner’s margin in each state is not always outside of the survey’s margin of sampling error. Rather than show states where the results are inside of the margin of sampling error as “leaning” or “toss-ups,” SurveyUSA for this illustration assigned Electoral Votes to the candidate with the larger share of the vote, no matter how small the winner’s margin. The Democratic nominee is not yet known. Running mates on neither side are known. These are not surveys of likely voters, these are surveys of registered voters.

Indeed, some of these states have undecideds that far outstrip the margins of error. Michigan, for instance, has Obama beating McCain only 46-45 and New Hampshire 46-44. Nevada goes Obama 46-41, with apparently 13% undecided. Virginia is 47-all, but the pollster gave the state to Obama.

It does give a handy baseline for people to consider when predicting how states will fall. However, the use of registered voters rather than likely voters diminishes the predictive value of this exercise. Republicans tend to be better about coming to the polls, which makes Democratic strength somewhat overstated in polls that don’t focus on likely voters. Survey USA at least admits these limitations, which is a refreshing change.

We have plenty of time for opinions to change. The electoral math will almost certainly change along with it.