If you followed the last presidential election, you saw how desperate people can get trying to predict the outcome of an election. You've heard analysts talk about polls and surveys. Ever wonder what methods are used by these analysts to dissect all the data that is out there?
The Election Prediction Game
Let's talk about one analyst in particular. Nate Silver, a baseball-statistics analyst turned election data-cruncher, became famous for correctly predicting the winner of the 2012 presidential election in all 50 states and the District of Columbia. How did he predict so accurately?
Silver built a new mathematical forecasting model that worked better than other established methods. He took all the polling numbers that were published every day, weighted them according to how reliable he thought they were, adjusted them to reflect other relevant factors, and converted them into projected vote shares at the state and national level. His model then simulated a large number of elections using a random number generator to determine vote tallies. This statistical methodology, known as Monte Carlo simulation, is widely used in many disciplines, including physics, finance, and engineering. It allowed him to correctly predict what would happen in the election!
Read more about Nate Silver's prediction here:
Use the links below to learn more about survey-based political forecasting methodologies.