Election Analytics website getting traffic as voting nears

10/30/2012 9:05:00 AM

With less than a week to go before Election Day, and what has been described as a "dead heat" between the Republican and Democratic presidential contenders, the Election Analytics website is seeing a lot of traffic.

Sheldon Jacobson
Sheldon Jacobson
"The election web site is doing great with around 62,000 views so far, and climbing. I think 100,000 are within reach," explained Sheldon H. Jacobson, director of the Simulation & Optimization Lab in the Department of Computer Science at Illinois.

"The Election Analytics website considers the Electoral College combinatorial chaos that will lead to each candidate winning the election,” Jacobson explained. “Updated daily, the site, which involves students in computer science and in industrial engineering, provides a nonpartisan snapshot of the election results.

"Prior to the first presidential debate, the election appeared like it would be a landslide. Since then, Governor Romney has been moving forward in the polls, making this race potentially as close as the 2000 race between Gore and Bush," Jacobson said.

Using the mathematical model that successfully predicted the outcome of the 2008 Presidential Election, the Election Analytics project employs Bayesian estimators that use available state poll results to determine the probability that each presidential candidate will win each of the states. These state-by-state probabilities are then used in a dynamic programming algorithm to determine a probability distribution for the number of Electoral College votes that each candidate will win in the 2012 presidential election.

For the 2012 elections, the Election Analytics Team has extended the model to handle both Senate and House races in addition to the presidential race. In the case of the Senate and House races, the individual candidate probabilities are used to determine the number of seats that each party will control. New prediction results will be posted as new polling data is made available.

Jacobson is well-known for his work in Operations Research (OR) that combines a multitude of tools in engineering, computer science, mathematical modeling and algorithms, and statistical analysis. It also utilizes tactics like probability theory, game theory, decision analysis, and simulation to find optimal solutions to complex problems.

OR’s countless applications have led Jacobson and his group to conduct research in areas as far-flung as aviation security, vaccine and immunization, public health, automobile fuel consumption, and NCAA basketball tournament brackets, as well as predictions for three previous presidential elections.

The ideas and methods used on the Election Analytics website originate in this paper written by Steven E. Rigdon, Sheldon H. Jacobson, Wendy K. Tam Cho, Edward C. Sewell, and Christopher J. Rigdon. A second paper provides an analysis of the prediction results from the 2008 Presidential Election.

Contact: Sheldon Jacobson, Department of Computer Science, 217/244-7275.

Photo: L. Brian Stauffer

If you have any questions about the College of Engineering, or other story ideas, contact Rick Kubetz, writer/editor, Engineering Communications Office, University of Illinois at Urbana-Champaign, 217/244-7716.