Obama won the election, as you well know. What you may not know is the interesting sub-plot that took place amidst the mathematical and statistical geniuses of the political realm: Who could most accurately predict the presidential election? Not just the winner, but the winners in each states along with the popular vote.
Would it be the respected organizations such as Mason-Dixon and SurveyUSA or would a relatively unknown statistical prodigy make it big? The answer turned out to be neither. Nate Silver most accurately predicted the electoral vote and the popular vote for the second presidential race in a row.
Silver got his start in the baseball world. He developed a program called PECOTA (Player Empirical Comparison and Optimization Test Algorithm) which could predict the career trajectory of a young baseball prospect. He sold PECOTA to Baseball Prospectus, a scouting network.
The baseball statistician has written several books, most recently, The Signal and the Noise: Why So Many Predictions Fail-but Some Don’t. Before that, Silver had primarily written or co-written books that had to do with baseball statistics. Think of the book (or movie) Moneyball. Silver’s previous books flowed in the same vein.
However, Silver’s career took a turn when he entered the political world by starting a blog called Fivethirtyeight which bears the tagline “Nate Silver’s Political Calculus.” Silver has garnered a staff that does extensive statistical analysis on various political races. An example of a current article is “Gay Vote Seen as Crucial in Obama’s Victory” where writer Micah Cohen analyzes how big a role the gay community played in re-electing the president.
Silver also continues to write extensively for fivethirtyeight. He also occasionally forays back into the baseball field or that of other sports arenas and writes statistical analyses such as “The Statistical Case Against Cabrera for M.V.P.,” but nearly all the articles written are politically driven.
Fivethirtyeight gained a considerable following after Silver accurately predicted 49 out of the 50 states vote in the 2008 Obama vs McCain Election. His formula outshone those of the more recognized political pundits and skyrocketed himself from sparsely known baseball analyst to widely-read political genius.
Fivethirtyeight and it’s genius Nate Silver managed to out-do themselves in the 2012 election. They correctly predicted all 50 states plus The District of Columbia and managed to determine the popular vote to within half a percentage point.
Some have said that Silver is not in fact the mathematical genius of politics that many claim him to be, but that he simply averaged all the reputable polls he could find. Other organizations, such as RealClearPolitics also predicted results that were very similar to Silver’s. However, Silver’s ingenuity comes into his understandings of probability. He, along with most that followed the election, understood that Ohio, along with the other “battleground” states, was pivotal for the election. Accordingly, Silver assigned risk factors to each of these states that meant the whole prediction wouldn’t tank with the loss of a key state.
Silver and Fivethirtyeight have proved that they are the pollsters to beat in the 2016 election. They have all proved the value of polling accurately and often. Now that the secret is out, will others rush to copy Silver’s formula or opt for other variables entirely?
~Caleb Morris