Going to the website 270towin.com and awarding each toss-up state to the leader of that state's poll average from Real Clear Politics, I came up with this Electoral College prediction:
Obama: 303
Romney: 235
That's Nevada, Colorado, Iowa, Wisconsin, Michigan, Pennsylvania (which isn't a swing state), Ohio, and Virginia for Obama and North Carolina and Florida for Romney.Virginia is the closest at +0.3 for Obama, but even then it would only make it 290-248 for Obama.
Entering this exercise I didn't think either could reach 300. Both candidates are uninspiring on their own merits and the enthusiasm for the partisans of each side seems to be more motivated by antipathy toward the other side. 303 seems a little high to me, and fallibility may strike me yet again, but at just about 5:45 p.m. here in the Central Time Zone I'll stick with 303-235.
There's a chance the polls will be wrong, but I wouldn't count on almost all the polls being wrong. There have been a few outliers in just about every state calling it a statistical tie or a slim Romney lead, most pretty consistently show Obama pulling ahead just enough in most of the states that will decide the election. That's why I suspect Romney has made a last-minute effort for Pennsylvania. It is my hunch that he knows Ohio is lost and needs to make it up somewhere. Unfortunately for Romney, I don't see him winner either.
Polling firms may or may not have a bias against Romney (or against Obama in the case of Rasmussen), but they don't have a good reason to be so. People don't trust polls just for the sake of it. If they were frequently wrong or too biased to be reliable nobody would pay any attention to them. We trust them because they are a reasonably accurate gauge of opinion and trends. As Robert Wright explains in this post at The Atlantic, we're talking about thousands of respondents in these polls, so they have a chance to get the random sampling that makes a poll reliable. 2012 has been the most heavily polled election in history and while one or two might be wrong, which is why Romney winning Virginia wouldn't surprise me, the chances that most would be wrong across several states is unlikely.
Get ready for a second term.