My own theory on predicting the 2008 race is pretty simple: I like FiveThirtyEight.com's credibility, but I want to work with the theory that Nate Silver is basically underestimating Obama's and the Democrats' margins by 2% across the board. 

This super-sophisticated theory is based on the following assumptions: (a) a high emphasis on Obama's superior ground-game which isn't built-in to the polling data, (b) accounting for a late surge not captured in the polls; (c) accounting for the early voting edge for Obama; (d) the cellphone advantage; (e) the Reverse Badley effect; (f) backlash to this weekend's Rev. Wright ads tipping some undecided voters into Obama's column, and (g) a wild ass guess/wishful thinking.

What would it mean if the Democrats overperform by 2%? At the end of the day, Obama will probably lead in the popular vote total of 53% to 45% (actually that's +1 for Obama and -1 for McCain over FiveThirtyEight.com's projections).

So what would the electoral map look like if Obama outperforms Monday's state polls (as interpreted by Nate's projections) by 2%?

Obama: North Carolina, Missouri, North Dakota, Indiana, Florida, Ohio, Nevada, Colorado, Pennsylvania, New Mexico, New Hampshire

McCain: Georgia, Montana, Arizona, Nebraska

Electoral Vote:
Obama: 378, McCain: 160

When I add 2% to the three Senate races leaning GOP, Chambliss (R), McConnell (R), and Wicker (R) still win. Therefore, I'm predicting 58 Dems (including independents), 42 Republicans in the Senate.  

I haven't calculated the impact of a 2% overperformance on House races, but a wild guess:
Democrats: 266, Republicans: 169.

Finally, I'm going with this prediction for the Minnesota Senate:

Franken: 42%, Coleman: 40%, Barkley: 13%