Election Day Explained
With the election today, it is easy to feel overwhelmed by the sheer volume of analysis, statistics and predictions coming from all sides. Spend any amount of time on Twitter, Facebook or a news site and you will hear everything ranging from an Obama landslide to a comfortable Romney win, by way of an electoral college tie. Even the most informed observers of the election appear to put forward wildly different opinions on how today will go, as pollsters contradict each other’s numbers, pundits’ predictions clash and campaigns accuse polling data of being ideological and ‘skewed’. Hopefully this post will go some way towards clearing up some confusion and give non-Americans in particular a better idea of what to expect over the next twenty-four hours.
A main source of confusion in recent weeks has been how informed people can hold such divergent views on how the election will pan out. For example, Nate Silver of the New York Times’ 538 blog recently offered to bet NBC’s Joe Scarborough $2000 (to be paid to charity) that Barack Obama would win the election, after Scarborough had called Silver a partisan ‘joke’ for putting the President so far ahead when polls showed the candidates to be neck-and-neck. In a sense Scarborough was right; the Real Clear Politics average of polls currently puts Obama just 0.7% ahead of Romney, but in relying on the national polling average, he was missing the point. In the American system, the candidate that gains the most votes overall does not necessarily win the presidency. In fact there have been four elections in which the victor has lost the popular vote, in 1824, 1876, 1888, and most famously in 2000, which saw Al Gore receive over 540,000 more votes than George W. Bush and still not take the White House. What is important in the American system is electoral votes. Each state is assigned a certain number (see the map above) and the candidate who wins the state takes those votes. Gain 270 or above, and win the election. The winner-take-all nature of these electoral votes is what is behind the disparity between the national polls and the actual probability of an Obama win – Romney’s 20 to 30 point lead in certain anti-Obama states is artificially inflating his level of support nationwide.
In reality, there are very few states that could go either way, and even fewer that would qualify as ‘tossups’ in the election. Places like California, New York and Illinois are rated as 100% certainties for an Obama win, while Romney counts Texas, Utah and Alabama amongst his ‘definites’. In fact, if all of the ‘safe states’ for either candidate are taken away, there are probably only nine remaining that are competitive; Colorado, Florida, Iowa, Nevada, New Hampshire, North Carolina, Ohio, Virginia and Wisconsin (and even from these we could probably give Nevada to Obama and North Carolina to Romney straight away). The New York Times has produced a wonderful interactive guide to demonstrate how each candidate could win the election, should these states fall one way or the other. The important thing to note in this guide is that of the 512 possible permutations they have found, 431 of them produce an Obama win, which at 84% is very close to Nate Silver’s most recent assessment of the likelihood of an Obama victory.
Regular readers of this blog may have noticed that I quote Nate Silver on quite a regular basis; his 538 model of election forecasting has proved repeatedly to analyse the aforementioned mass of data with remarkable accuracy. In 2008, his model correctly predicted 49 of the 50 states in the presidential election, as well as all 35 Senate races that year. However, in recent days he has come under fire from conservative critics such as Joe Scarborough, for giving Barack Obama too great a chance of winning. Certain pundits have gone so far as to suggest that he should resign in disgrace should Mitt Romney take the presidency, as he and his model would be undermined. This originates in two places; first, no pundit wants to say that one candidate has a 90% chance of winning. Pundits love a horse race, because it validates their existence, and brings viewers, blog hits and newspaper sales. If the race is over by August, these pundits will not have much to do until November. The other point is that people do not understand the difference between a prediction and a projection. When Scarborough launched his attack on Silver, the 538 forecast put Obama’s chances of re-election at 73%. Scarborough found this ridiculous and grounds to discredit Silver if Romney won. But imagine a jar of 100 coloured balls, 73 of them blue and 27 red – would you be shocked to pull a red ball out of that jar? Even now, when Silver puts Obama’s chances at 91%, you could still take a red ball out of that jar, and Mitt Romney could still win. It just is not as likely, and there is no reason to throw all belief in statistical probability out of the window if it happens.
To return to that wonderful New York Times ‘512 paths to the White House‘ tool I mentioned above, you may notice that alongside the 431 ways Obama could win and the 76 ways Romney could win are 5 ways in which the election could end up tied, 269-269. While this is a very unlikely eventuality, it certainly could happen and if it does, the American system has a way of resolving such a deadlock. First, the newly elected House of Representatives would vote to choose a president from the candidates (with each state receiving one vote), and then the Senate would choose a vice president from their running mates. Of course as the House has a Republican majority and the Senate has a Democratic majority (and both are expected to stay this way), this would end up with a President Romney with a Vice President Biden. This eventuality, with a divided White House and the 53-strong majority-Democrat California delegation receiving the same number of votes as the one (Republican) representative from Wyoming, would undermine the office and the country to no end.
So my advice to supporters of both sides today is this; ignore the national polling, which means very little. Stop attacking people like Nate Silver or demanding that they be all-knowing wizards, because they’re just attempting to demonstrate the most likely scenario, without necessarily claiming it to be inevitable. And most of all, whoever you support, pray for a clean win. In the current climate, the last thing anybody needs is an acrimonious battle through the courts, with recounts and claims of illegitimacy.
And one more thing. If you have a vote, use it. Because there are a lot of people like me around the world who would love to have a say in this election. So don’t waste your voice.