Election 2012 thoughts
Just a few days away until Election 2012! If you are reading this then, like me, your pulse probably quickens a bit just thinking about it. Unlike most of my blog posts in which I address a particular theme at length, I am going to simply go on-the-blog-record with a few thoughts about Tuesday.
1). Hurricane Sandy was a catastrophic event that took a hundred lives and cost tens of billions dollars worth of damage. It also almost certainly also helped President Obama, whose performance during and after the storm registered the approval of almost eighty percent of Americans. It’s hard to imagine a bigger boost for Obama’s election prospects the week before the election. If the hurricane does make a difference for Obama in the election and an Obama victory decides the fate of our new universal health care and the estimated 20,000 plus lives a year it will save, then Sandy will end up sparing many times more lives than it took.
2). Even before the hurricane, President Obama had been the clear favorite to win the only vote that matters — that of the Electoral College. The mathematics are in the President’s favor because he can afford to lose some, even most, swing states, and still win the election. Romney, on the other hand, has to win almost every swing state.
Taking a look at George W. Bush’s reelection map from 2004, it becomes clear in hindsight that the former president basically ran-the-table on John Kerry, and still won by only 34 electoral votes. With New Mexico having become a strong, largely uncontested blue state since then, that’s a net swing of 10 more votes toward Obama. Comparing 2012 to 2004, Obama only has to hold on to the states that Kerry won, plus win Ohio OR Virginia OR Florida OR lose all three but win combination of 2-3 other smaller swing states in order to win the election.
3). There is a real possibility that President Obama wins the electoral vote and loses the popular vote (but not vice-versa). National polls show a dead heat between the two candidates, but state polling continues to show a fairly clear Obama electoral victory.
Is a divided electoral college-popular vote the likely outcome? No. As Nate Silver pointed out this week, a meta-analysis that combines state polls shows a slightly more favorable national outcome for President Obama than random national sampling. Furthermore, combining state polling in this manner has been more accurate in the past than national polling.
Still, what would happen if there were a Romney popular vote victory and electoral loss? Would Democrats take satisfaction in taking revenge for the 2000 election? Sure – but hopefully, were it to happen, it would also spark a bipartisan effort to amend the Constitution and get rid of the absurd Electoral College system.
4). The polls might be biased, but not intentionally. When we think of the word bias, we think of racial bias and other similar character flaws. In polling, however, bias resides in the assumptions that pollsters make about response rates to their polls and whether those polled will actually vote. Every serious poll engages in stratified sampling techniques that make assumptions correcting for the fact that, for instance, men might answer the phone more than women and that older people might be more likely to vote than younger people. Contrary to what many pundits say about pollsters having an agenda, polling companies want to get the underlying assumptions right because they have a lot of money and a lot of reputation on the line.
Nevertheless, Romney’s best chance to win the election is for there to be an unknown bias at work in most of the polls that have been taken in recent weeks — a mistaken groupthink concerning response rates or turnout that would leave everyone scratching their heads on Wednesday.
5). A few strange things could still happen, even if there is not much time before the election. Although unlikely, President Obama could be caught saying something really unpopular about, say, guns and religion. Hopefully (and certainly) even more unlikely, there could be a terrorist attack that influences the outcome. There is no week that “Al Qaeda types” would like to strike more than the week before a US Presidential election. Given the political impact that the Benghazi attacks have had, the incentive is that much greater.
Weather conditions less dramatic than a hurricane could even affect the election. Bad weather depresses Democratic more than Republican turnout. Democrats should be pulling for some beautiful autumn weather across the country — or at least in the swing states.
So, what’s my prediction for the election? Obama wins 303-235 in the Electoral College and wins the popular vote 51.5-48.5%. Any of you have your own prediction?