Karl Rove: For Romney to Win "Some Polls Have to Be Wrong"

| Tue Nov. 6, 2012 12:50 AM EST

Even Karl Rove, the political genius of the Republican Party, admits it: For Mitt Romney to win the presidential election on Tuesday, "some polls" have to be wrong. That's because those polls, especially in key swing states, show President Obama headed for victory, albeit a very narrow one.

Rove went on Fox News Monday night to give his final assessment of the Obama-Romney showdown. His prediction was that Romney would scrape together a win with 285 electoral votes, all but sweeping the president in the key swing states of Colorado, Florida, Ohio, North Carolina, Virginia, New Hampshire, and Iowa. And while Rove's own analysis of recent presidential polls put the presidential race at a dead heat, he conceded to Fox's Bret Baier that "some polls have to be wrong a little" for Romney to win.

FiveThirtyEight's Nate Silver is far less charitable toward Romney: He predicts Romney will claim a miserly 224 electoral voters, and stands only an 8.6 percent chance of winning the election. The right-leaning RealClearPolitics says the race is a statistical tie at 48 percent for both candidates.

Here's the video of Rove (the segment starts at the 3:30 mark) followed by a transcript of the moment in question:

Baier: Bottom line: For Romney to win Tuesday, these polls have to be wrong.

Rove: Some polls have to be wrong a little, because the race is that close. Remember, take a look at the national polls. Just simply in the last week, 23 polls, you average them all together, 48.3 [percent] for Romney, 48.1 for Obama. That's as of 10 o'clock this morning. So it is dead even, knife's edge, long night, exciting outcome.

The way all the polls look, Rove's likely to be right about one thing: It's going to be a long night on Tuesday.

Get Mother Jones by Email - Free. Like what you're reading? Get the best of MoJo three times a week.