Nate Silver may have a pretty good track record, but I pretty sure he'll be shown wrong this time. There are many factors one can look at (and I have read Nate's analysis), but I think like many he overestimates the likely democrat turnout, and overestimates other factors. On the other hand, I think the Overskewedpolls guy may go too far in the other direction. I think it can be boiled down to this. Most polls still show a tight race, with Romney a little ahead in the popular vote, and Obama having a small lead in most swing states. From this alone one could argue for a split popular/EC result with Obama the winner. However, most of these polls are still oversampling democrats 4 to 8 points. So, in order for Obama to win, he would have to have a democrat turnout in his favor of at least that margin, even assuming no party cross-over (more on that below). But several lines of evidence (including some polls specifically on this question, and trends from
2010), suggest that the turnout will be split fairly evenly dem/rep, if not a point or two +R. That alone could clinch it for Romney. On top of that, Romney leads by double digits among independents, and more dem/rep crossover is expected in his direction than the other way around. I think all this spells a Romney victory in both EC and popular counts. Both sides have tried to argue that early voting trends (based on rep/dem ratios or polls, not actual vote counts) are favoring their side. So the data so far is limited and hard to judge, and most I've seen is compatible with the above analysis, especially when you factor in the historically heavy republican over democrat turnout on election day.
Obama supporter David Axelrod says he'll shave his mustache if Romney wins PA. It's interesting he didn't bet his facial hair on the whole election, but I will. I'll bet my beard against against anyone elses 'stach or beard that Romney will win the election. For any women or clean shaven men who want in on this, I'll bet a large pizza instead.