The polls are looking favorable for Obama going into the final weekend before the election.
From the battleground states:
Colorado: Obama 46%, Romney 46% (Reuters/Ipsos)
Colorado: Obama 50%, Romney 46% (Public Policy Polling)
Colorado: Obama 47%, Romney 45% (Denver Post/SurveyUSA)
Florida: Obama 48%, Romney 46% (Reuters/Ipsos)
Iowa: Obama 49%, Romney 45% (Gravis)
Michigan: Obama 52%, Romney 47% (Rasmussen)
Michigan: Obama 52%, Romney 46% (Public Policy Polling)
Nevada: Obama 50%, Romney 44% (Mellman)
New Hampshire: Obama 50%, Romney 44% (New England College)
New Hampshire: Obama 50%, Romney 49% (Gravis)
Ohio: Obama 50%, Romney 47% (CNN/ORC)
Ohio: Obama 49%, Romney 49% (Rasmussen)
Ohio: Obama 47%, Romney 45% (Reuters/Ipsos)
Ohio: Obama 50%, Romney 46% (We Ask America)
Virginia: Obama 48%, Romney 45% (Reuters/Ipsos)
Virginia: Obama 49%, Romney 48% (We Ask America)
Wisconsin: Obama 52%, Romney 45% (We Ask America)
ABC News/Washington Post: Obama 49%, Romney 48%
Public Policy Polling: Obama 49%, Romney 48%
Purple Strategies: Obama 47%, Romney 46%
Rasmussen: Obama 48%, Romney 48%
Reuters/Ipsos: Obama 46%, Romney 46%
Romney could still win, but would have to out-perform the polls by over two percent to have a chance. The Denver Post has nine electoral college predictions–showing different combinations of states which lead to an Obama victory.
Supporters of each party are looking for ways in which their party could out-perform the polls (with Obama merely needing to match the polls at this point). Both parties have argued that early voting is helping them. The problem for the Republicans is that much of their early voting is occurring in southern states which will go Republican regardless of when people vote. The real question is not who is getting the most early votes, but whether Democrats will increase their total turnout with early voting. Polls of all registered voters typically show the Democrats doing five points better than polls of likely voters. If the Democrats can narrow this gap they can boost the numbers above.
Back in 2004 liberal blogs were counting on the Incumbent Rule to give Kerry the victory. The basic idea is that if the incumbent is running at under 50 percent, the majority of undecided voters will break for the challenger (already knowing the incumbent), giving a challenger who is close behind the victory. That didn’t work for Kerry, and it doesn’t look like this will work for Romney.
Other factors might also alter the results compared to the polls. The Libertarian Party, along with the Constitution Party in Virginia, might take a small number of votes away from Romney. I don’t see the Green Party as being a threat to Obama this year as Nader was to Al Gore in 2000. The Constitution Party’s candidate, Virgil Goode, is from Virginia and has the potential of taking enough votes from Romney to give Obama the state in a close race, while Libertarian Party candidate Gary Johnson might be a spoiler in some western battle ground states.
There is speculation that the polls might be under-counting Latino votes, possibly enabling Obama to do several points better in some states, as Harry Reid did when running for reelection two years ago.
Under counting cell phone users might also play a part. Polls using robocalls are legally not allowed to call cell phone, underestimating younger voters who are more likely to vote Democratic (assuming they do show up to vote). Polls not using cell phones do try to adjust their numbers but at least one Democratic pollster believes that Obama is actually doing much better than the polls show.
These factors favor Obama, and there is one more trend which helps Obama. He had the far better week, denying Romney the chance to regain the momentum he held after the first debate. Besides just dominating the news, he benefits from comments from Chris Christie, the endorsement from Michael Bloomberg, and the report of an increase in jobs created. There is very little time left for something to happen to change the trajectory of the race.