If you have limited time to devote to following the presidential election this fall, I suggest you follow the data on just three states: Florida, Virginia, and Ohio. Mitt Romney's only likely path to victory over Barack Obama is to win those three states. Conversely, if Obama can carry just one of those states, he will likely win the election.
All of the predictions in this column update in real-time on PredictWise. I create the predictions with a combination of three types of data: polls of voter intention (via the HuffPost Pollster API), prediction markets, and fundamental data. For the unfamiliar, prediction markets, such as Betfair and Intrade, are markets where the user buy and sell contracts on outcome of the upcoming election.
Obama is extremely likely to carry 191 electoral votes from 15 states and DC. Likewise, Romney is extremely likely to carry 167 electoral votes from 20 states. In these states, the other candidate has a negligible likelihood of flipping the state.
Obama has 46 electoral votes from three additional states where Romney has a non-negligible likelihood of stealing a state (Obama's likelihood of victory): MN (89.5 percent), PA (89.2 percent), and MI (86.6 percent). Romney has 39 electoral votes from four additional states where Obama has a non-negligible likelihood of stealing a state (Romney's likelihood of victory): AZ (94.4 percent), MO (90.3 percent), ND (86.9 percent), and NC (85.2 percent).
I actually expect, on average, one of these states to flip. Electoral College elections are not independent outcomes; it is highly likely that any candidate that picks up a state where he had a 5 to 20 percent likelihood of victory has also won a lot of states where he had a 20 to 50 percent likelihood of victory. In order to capture a long-shot state, a candidate needs something more than an idiosyncratic shock to that state, but also a national trend that carries a few other states with it as well. Thus, it is extremely unlikely that the long-shot state would swing the election; rather, it will pad a solid victory.
Before considering the only eight swing states, Obama has 237 electoral votes to Romney's 206 electoral votes; the winner needs 270 electoral votes or 33 more for Obama and 64 more for Romney.
Florida (29), Virginia (13), and Ohio (18) account for 60 or the 95 remaining electoral votes and constitute Romney's only viable path towards victory. Victory in all three would require Romney to just pick of one of the other five remaining states (Obama's likelihood of victory): Iowa (52.1 percent), Colorado (60.8 percent), Wisconsin (67.5 percent), New Hampshire (71.2 percent), and Nevada (73.7 percent).
If Obama wins Florida (41.1 percent), there is almost no likelihood that Romney sweeps the rest of the states to take the presidency. This is the only swing state that leans Romney. If Obama wins just Virginia (52.4 percent) or Ohio (58.9 percent), it is possible for Romney to win, but highly unlikely, because he would have to virtually sweep the remaining swing states.
Where does this leave the overall election for me as we head into the Republican National Convention -- I have Obama with a 58.9 percent likelihood of reelection, or the exact same likelihood as his chances in the pivotal swing state of Ohio.