When I first began this blog, my idea was to focus on enduring issues rather than on current events. I’ve broken that rule already with a couple of posts dedicated to analyzing the Presidential debates. There’s a bit of an irresistible spiral that occurs when you begin blogging on politics – you start reading more, and watching more, and the temptation becomes strong to comment on everything. I presume I will have to blog an official prediction prior to election day, since I’ve been addicted lately to following various poll-based electoral maps. As a prelude to that, I’ve decided to dedicate a post to some of the various scenarios.
I have been relying heavily on two websites, each of which has some unique and helpful features in analyzing possible electoral scenarios:
I’ve been following electoral-vote.com most closely. The site is put together by a person who claims to be a libertarian that leans Democrat, but regardless of any particular personal preferences, appears to be scrupulous about providing up to date, accurate polling data. At the main page, visitors are greeted by a beautiful electoral map, with states colored 3 shades of blue or red depending on whether recent polls average strongly, moderately, or barely in favor of President Obama or Mitt Romney. Below the map is a list of states whose numbers have been updated that morning with a new poll. Hovering the mouse over any state gives the polling average for each candidate and the polls on which that average is based. The site also makes it incredibly easy to scroll back in time so that you can analyze polling trends.
Real Clear Politics is a portal for all things political, collecting links to editorials and news stories from all corners of the ideological world. For electoral mapping, the site is more confusingly organized than electoral-vote.com. Like that site, the electoral map on the RCP page has 3 levels of color coding representing the likelihood of President Obama or Mitt Romney winning each state, again based on averages of recent polls. I have not been able to find a clear description of how the states are color-coded on the RCP site, but one difference between RCP and electoral-vote.com is that RCP has many states colored gray and listed as a Toss Up. Whether these are states with polls within a statistical margin of error, or whether the RCP site used some cut off (e.g., a 5% difference) is not clear. The fun on the RCP page comes when you press the Create Your Own Map button, which allows you to test the effect of changing the status of a state, for example, from Toss Up to Leaning Romney or Leaning Obama.
If you compare electoral-vote.com’s map for today (October 25) to the map at the beginning of the month and before any of the debates (October 1), it is clear that the race has tightened and that all of the momentum favors the challenger Romney. The site has the President holding a 294-244 vote lead, compared to 347-191 on October 1. Florida, North Carolina, and Colorado shifted in that period of time, and other states, some considered reliably Democratic, have tightened (Wisconsin, Nevada, and New Hampshire).
What’s troubling if you are a Romney fan is Ohio. Ohio is one of the few close states in which registered Republicans outnumber registered Democrats. Ohio is a state whose voter demographics mirror the country as a whole. The state elected a Republican governor in 2010, John Kasich, whose policies have arguably helped to improve Ohio’s economy. Yet Ohio seems to have stubbornly resisted the Romney momentum that has affected so many other states. The 50-44 advantage the President held on October 1 has only narrowed to 48-46 in the most recent sample (from electoral-vote.com). And Ohio has been one of the most heavily sampled states in the country, given that many people believe Romney can only win by winning Ohio.
RCP’s numbers are identical. The last 9 Ohio polls RCP provides score the race as a tie in 3 cases and an Obama lead (ranging from 1 to 5 points) in 6 cases. The margins are always razor-thin, but there are two kinds of razor-thin in polling. One kind is razor-thin because one guy is ahead in two-thirds of the polls, and the other guy’s ahead in one-third of the polls. The other kind of razor thin is because one guy’s got a narrow margin in all of the polls. That’s the case in Ohio, and that’s bad news for Romney. Or to put it more accurately, Romney is ahead in none of the polls. To add insult to injury, Romney may have less hope for a change in Ohio than in other states, since some have estimated that about a third of Ohio’s voters have already cast ballots in early voting.
I’ll save my predictions for later, but one that seems hard for me to imagine changing is that President Obama will win Ohio. A corollary to that prediction is that I believe it is unlikely that Romney will win the election.
One other trend worth mentioning is that Romney has shown encouraging signs in unlikely places: Wisconsin, Minnesota, Pennsylvania, Michigan, and New Hampshire. A Romney fan might hope for an upset in one of these states, though my best guess is that the best result won’t be winning one of these Democratic states, but rather coming closer to winning than expected. (The adage that close only counts in horseshoes is relevant here.) It seems logical to me that the upsets most likely obtained among these states would be in the states with the least population, since an advantage in voter enthusiasm – which Romney now appears to have – would most easily make a difference in the smaller states. Of course the corollary to that is that the upsets would most likely come in the states that mean less in the electoral math – New Hampshire, not Pennsylvania; Minnesota, not Michigan.
Does Romney Have a Path To Victory?
Assuming that Romney loses Ohio, and placing low odds on Romney pulling off an upset in a populous traditionally-Democrat state despite showing some Romney momentum, makes the picture sound bleak for Romney.
But consider the following scenario. As of today, the RCP electoral map shows a 281-257 lead for President Obama using the No Toss Up map. The No Toss Up map assumes that each state will fall to whomever has the higher average polling numbers, regardless of margin of error or any “too close to call” threshold. I took a look at that map and mentally rated the blue states that I felt were most-likely to turn towards Romney.
The first state I selected was Nevada. My rationale for Nevada was the following. First, the economy of Nevada is horrible – the state has the highest unemployment rate in the nation. The economy is one issue on which Mitt Romney seems to have a clear advantage. Second, 11% of Nevada’s population is Mormon – it may be that Mormons who wouldn’t normally come out to vote might do so to cast a vote for the first Mormon President. One need only look back to 2008 to see a similar effect when African Americans turned out in record numbers. Third, Nevada is a low-population state with a small number of TV markets, one in which a dedicated advertising campaign might educate just enough voters to make a difference. Fourth, although the state went strongly for President Obama in 2008, it went for George W. Bush in 2000 and 2004.
The second state I selected was Iowa. Iowa tends to vote Democratic, but it did vote for George W. Bush in 2004. It is another small state with a small number of TV markets, and is certainly more conservative than other upper midwest states like Minnesota, Wisconsin, Michigan and Illinois. In some respects the state is more similar demographically to Indiana than these states, and Indiana is strongly in Romney’s camp this time around.
What happens to the electoral map if one nudges Nevada and Iowa into Romney’s camp? It’s an electoral tie: 269-269.
What happens in an electoral tie? The House of Representatives votes for President in a very unusual fashion. Each state casts one vote. So in the House vote, California, with 53 representatives, has the same clout as Alaska, with 1. We can predict that, since Republicans have a majority in 33 states, Mitt Romney would likely be elected President. That’s not for certain though, because the Senate is charged with selecting the Vice President, and the Senate delegations are predominantly Democratic. If President Obama were to win the popular vote in a 269-269 scenario, there might be enormous pressure on the House to strike some sort of unprecedented bargain. (Or not so unprecedented: see the elections of Rutherford B. Hayes, John Quincy Adams, and Thomas Jefferson.)
Best money though is an electoral tie goes to Mitt Romney. The Iowa/Nevada flip represents a plausible path to victory for Romney even without Ohio. Other scenarios are possible, such as Wisconsin and New Hampshire flipping, but these scenarios seem less likely.
Although I’ve made this path sound plausible, there is still a very big hidden assumption. The No Toss Up map comfortingly blankets a lot of the country in red, but many of these states are a light pink at best. Thus, Florida, Colorado, Virginia, and North Carolina are listed as Toss Up states at the moment, but they can’t be for this pathway to work. Any one of these states – or all of them – could end up in President Obama’s column, as all of them were in 2008. Romney can only lose some of these states (not Florida), and then must replace the lost state with Ohio (or some less likely state, like Minnesota or Pennsylvania – and Minnesota would only replace Colorado). If I had to bet today, I would give Florida and North Carolina to Romney, but Colorado is a true coin flip and Virginia has been frustratingly resistant to returning to the Republican camp since McCain lost the state for Republicans in 2008. Prior to 2008, Virginia last went Democratic in Lyndon Johnson’s 1964 landslide.
Of course, tomorrow is another day. We’ve got two more weeks to play around with these silly red and blue maps, and then life gets back to normal (barring the torture that followed the 2000 election). Right now, I’d say, don’t turn off your TVs if Ohio comes back blue. But if Iowa and Ohio turn out blue, I think it will be 4 more years for the current President of the United States.