We’re 29 days away from one of the most anticipated elections of all time. Some experts call it the most important election since 1860.
While we can’t back that up with real data analytics, what we can tell you is that this is the most data-driven election of all time.
Let’s start with some information from a guy who predicted the winner in 49 out of 50 states back in 2008.
Ever heard of Nate Silver? He’s developed a Major League Baseball career ratings system and predicted the winner of the 2008 election. Additionally, he’s created a NY Times-licensed blog called “Nate Silver’s Political Calculus.”
This statistician is currently predicting a win for the incumbent, President Barack Obama, and he’s developed his prediction model for the last election based on sabermetrics.
According to Silver’s model, as of last night, the president has a 78.4% chance of winning the election. You can find real-time updates at Silver’s blog FiveThirtyEight.com.
Another model you may have heard of is the 13 keys model made famous by historian Allan Litchman. His predictions are based on 13 questions that can be answered either “true” or “false.” The “true” answers favor the incumbent party. If the answer to five or fewer questions is “false,” then the incumbent party retains the presidency; if the answer to six or more is “false,” the challenger wins.
The 13 keys model is based on a statistical model, but it’s a popular prediction model because of its simple presentation. And it hasn’t been wrong since 1984.
Litchman predicted the president’s win a year ago. He’s added a few caveats for this year’s election, which are outlined in the September/October issue of Analytics Magazine. The only such caveat in play right now is the potential for major political or foreign policy failure to take place before the election.
“The key doesn’t turn because of strongly critical opinions by a few people, even very knowledgeable people, about certain polices and actions. Broad public perception is what counts,” writes Doug Samuelson, the author of the article in Analytics Magazine.
A third model that’s been developed by two political science professors from the University of Colorado predicts a Romney win. Professors Kenneth Bickers and Michael Berry predict the challenger will win 330 of the 538 Electoral College votes, according to a recent Times Call report. They base their forecast on state-by-state economic data such as the unemployment rate and changes in income.
So, as far as the data analytics of the election go – there are 13 prediction models that are outlined in the August issue of PS: Political Science & Politics. Five of the models predict the president, five predict Romney and three are up in the air, reports Brittany Anas for the Times Call.
Amanda Brandon
Spotfire Blogging Team
