George Will makes a remark in his column today that isn’t entirely accurate:

Self-identified conservatives outnumber self-identified liberals 2-to-1 in a nation that has reelected the most liberal president since Lyndon Johnson and his mentor Franklin Roosevelt.

Looking at national exit polling, the electorate that turned out this year doesn’t fit that description. I’ll leave the “most liberal since LBJ” designation for another time (just review the domestic records of Nixon and George W. Bush if you believe that). FoxNews’ national exit poll found that liberals made up 25% of the electorate, moderates approximately 40%, and conservatives 35%. Self-identified conservatives outnumbered liberals in the presidential vote by just ten points, which is a far cry from a two-to-one advantage. That’s consistent with what other organizations found.

Will’s 2-to-1 figure can be found in other surveys of ideological self-identification, and it’s true that a person’s self-identification is not always a reliable indicator how he votes or what policies he favors, but it doesn’t account for this election result. The 2012 electorate was slightly more liberal (25%) than the 2008 electorate (22%), while conservative numbers barely increased (35% instead of 34%). If we reviewed the exit polling and noted instead that an electorate that is reportedly 65% non-conservative cast most of its votes for the center-left candidate and his party, we would not be baffled or puzzled by the outcome. If all that conservatives had to do was outnumber liberals, a ten-point advantage might be enough to prevail on a regular basis. As it happens, there are still the other two-fifths of the population left over that don’t identify entirely with either. According to the exit polls, Obama won the moderate vote by 15 points.

I dwell on this because misreading the composition of the electorate was the main error that Will and others made in their arguments for what would happen in this election, and this misreading contributes to unnecessary puzzlement about why people voted as they did. Ignoring a large segment of the electorate (in this case, moderates) doesn’t add to one’s understanding of the political landscape, and consistently overstating conservative numbers and the conservative character of the electorate reinforces assumptions that virtually all that conservatives need to do to win elections is show up.