How might Romney win?

I have been very bearish on Mitt Romney's prospects to win the U.S. presidential election for quite some time.

Back in January, I suggested Barack Obama might even beat Romney in the deep south. This no longer seems plausible. I assumed that the Obama campaign and/or an aligned Super PAC would run ads throughout the south touting Romney's record as a liberal on gay rights and abortion, prompting much of the Republican coalition in the south to stay home or vote for a third party - likely Virgil Goode. By not doing this, the Democrats have assured that the deep south will remain an easy win for Republicans, thus giving Mitt Romney a shot at the presidency.

In the past few days, two of my favourite U.S. political blogs - The Fix and Five Thirty Eight - have both noted that there are few winning paths for Romney. This got me tinkering with electoral scenarios myself.

The Fix made an excellent point in pointing to George W. Bush's winning path in 2000. Bush won narrowly by 271-267 electoral votes, and that map more so than Bush's 2004 map seems a good template for Romney. However, due to population shifts that same map would yield a more comfortable 285-253 victory today.

The biggest holes in the 2000 Bush map for Romney are Ohio - which now seems all but certain to go for Obama - and Virginia which is very competitive. These are big ticket items with 18 and 13 electoral votes respectively. I had assumed that without them, Romney would be largely hopeless. However, I was surprised to see how many realistic paths to victory remain for Romney even if these two states come off of the table.

As above, if Romney matches the 2000 map he wins by 22 electoral votes.

Were he to lose Virginia, he would still win a narrow 272-266 victory. Were he to lose Ohio but win Virginia, he would need only add one of Iowa or Wisconsin (both plausible wins) to get over the top. Were he to lose both Virginia and Ohio while picking up both Wisconsin and Iowa, he would win 270-268. That implies quite a bit more flexibility than I imagined.

Mitt Romney remains a heavy underdog, but there are more realistic paths to victory for him than I thought. However, it seems that New Hampshire - down as a win for Romney in all of the above scenarios - despite its paltry 4 electoral votes is critical to his path to victory. If he cannot win Ohio or Virginia, those four votes become critical to almost all victory paths for Romney. It is no wonder that despite its small size, New Hampshire is listed as the 8th most likely tipping point state by Nate Silver at Five Thirty Eight.