CSUF Engineering Math Model Predicts Next US President

Chandrasekhar Putcha, professor emeritus of civil and environmental engineering at Cal State Fullerton, has been using math every four years since 2008 to answer the question everyone is asking: Who will be the next U.S. president?

In the race for the White House in 2024, Putcha's mathematical model predicts that former President Donald Trump will win with 47.95% of the vote and 271 electoral votes, while Vice President Kamala Harris will win 46.12% of the vote and 267 electoral votes – a 2% error rate. The winning candidate needs 270 electoral votes to win the presidency.

Putcha notes that the presidential race remains uncertain even a week before the election. National polls show that battleground states remain extremely close, with neither candidate having a clear lead in the states likely to decide the presidency.

In previous presidential elections, the mathematical model correctly predicted that Joe Biden would win the presidency in 2020 and Barack Obama would win the presidency in 2008 and 2012. In the 2016 presidential election between Trump and Democratic candidate Hillary Clinton, the model predicted Clinton to win the presidency.

Putcha said the model's calculations were correct in 2016, but the survey data used was incorrect.

“If the poll is accurate, the mathematical model should accurately predict the outcome of the presidential election,” Putcha said.

How does the technical mathematics model work?

The mathematical model uses technical principles based on probability and statistics to analyze survey data from various well-known sources, which Putcha says is sufficient for forecasting purposes. Putcha notes that the engineering math model is more robust than what a political scientist would use.

The methodology covers both the popular vote and the electoral college. The model uses election poll data from various sources – such as ActiVote, Bloomberg/Morning Consult, Dartmouth College, Emerson College, New York Times/Siena College, SurveyUSA, UC Berkeley's Institute of Governmental Studies and the Washington Post. It calculates the probability of winning for each candidate and assigns the corresponding electoral votes from each state to each candidate. The two sets of data are then combined to determine a winner.

Factors that could skew the model's prediction include undecided voters or voters who may refuse to participate in public polls, Putcha said.

“The main problem is that many people do not respond to calls to take the surveys,” Putcha said.

To help the model produce more accurate results, Putcha said this year the model includes an “uncertainty factor” to account for voters’ gender, ethnicity, age and race.

Putcha said his team calculates a 97.4% confidence level in the polls.

About the team

Putcha began teaching at CSUF in 1981 and retired in 2021. Over the course of his 40-year academic career, Putcha's interdisciplinary research focused on technical reliability and risk analysis. He teaches part-time at Cal Poly Pomona and continues to conduct research in risk analysis.

Putcha's research team includes CSUF alumnus and graduate student Vineet Penumarthy '16 (BS Mechanical Engineering) and economist Brian Sloboda.

View Putcha's 2024 Model Prediction presentation and access presentation slides and photos in Dropbox.