Nate Cohn of the New Republic explains why the President is still a 2 to 1 favorite to win reelection despite polling terribly with white voters. In short, there are just a lot fewer white voters in proportion to all other citizens. And the GOP strategy to bank on making immigration reform a wedge issue to energize white voters could have disastrous consequences for the party. From Cohn:
But the polls do not anticipate strong Latino turnout and Obama can make up ground if his ground game can upset expectations. Incredibly, polls suggest that Obama might do better among Latino voters than he did in 2008.There probably isn't anything the Republicans could have done to significantly improve their standing among black voters so long as they faced Obama, but there are plenty of Latino swing voters and 40 percent voted for Bush in 2004. If Romney makes a comeback in Ohio and Obama regenerates Latino turnout and wins Latino voters by as much or more than he did in '08 in states like Colorado, Nevada, or Florida, there's a chance we look back on the Republican decision to oppose comprehensive immigration reform (not to mention the DREAM Act) as the moment that ultimately cost them the 2012 election.
This prophecy seems to be playing itself out in Nevada. Remember, Democrat Harry Reid was the underdog the entire race in 2010 to retain his Senate seat. Pollsters and prognosticators, including some of my favorites like Nate Silver, consistently predicted his loss. But the pollsters underpolled Hispanics and also assumed they wouldn't vote in the numbers they did. He won as did other Democrats in states with large Latino populations. Remember, the polls now are counting two groups of people - those who are telling pollsters how they have voted and those who have yet to vote telling how they intend to vote. There is obviously no spread between registered voters and likely voters for the first group. And high turnout in early voting for a group of voters also tells us that the group is energized and reduces the spread between likely voters and registered voters.
17% of Nevadans have already cast their votes. And the Democrats have a lot about which to be happy. From the AP:
After four days of early voting, 17 percent of active Nevada voters have already cast ballots in the upcoming election.Figures released Wednesday by the secretary of state's office show 214,609 people have voted, either in person or by mail, since early voting began Saturday. Of those, ballots received from Democrats total 99,415, compared with 79,697 cast by Republicans. The remaining ballots are nonpartisans or voters registered with minor political parties.The roughly 173,000 people who have voted in person are nearly 43,000 more than those who voted during the same period in the 2008 presidential election.
There is no way to tell whether Democrats will turn out in large numbers throughout the process, but this is starting to look like a re-run of the Reid election. Furthermore, two new polls in Nevada are showing this early voting already helping the President. From pollster PPP:
Obama is already well on his way to winning Nevada based on early votes that have been cast in the week. Among those who say they've already voted he has a 61/39 advantage. Romney is up 51/46 with those planning to vote between now and election day.
Obama's leading based on advantages of 69/28 with Hispanics, 82/12 with African-Americans, 54/44 with women, and 58/39 with young voters. Romney's ahead with men (50/48), whites (57/42), and seniors (53/45). He also has a 53/44 advantage with independents but he'd probably need to take them by a wider margin than that to overcome the Democratic registration edge in the state.
Look for a similar story to play out in Colorado, Virginia, Florida and North Carolina. I'm not saying Romney won't win these states, but I would expect assumptions about the Latino vote to similarly give Republicans an inflated number in the polls. We'll learn more as we hear about how early voting is going.