With only days left until election night, Chandrasekhar Putcha, a Cal State Fullerton professor in the department of civil and environmental engineering, created a statistical method that predicted that the former vice-president and Democrat presidential candidate, Joe Biden will win the 2020 election.
According to the data, the mathematical model anticipated that Biden would win by a wide margin with 49.22% of the popular vote and 350 of the electoral votes. President Donald Trump will trail behind with 45.58% of the popular vote and stay behind in electoral votes with 188 votes.
With a research team of three people led by Putcha, along with Brian Sloboda, an economist from the University of Phoenix and Vineet Penumarthy, a CSUF graduate and design engineer, they mathematically compiled data to come to their conclusion.
“I will say the confidence that we have in this is a 95% confidence we have that we are right,” Putcha said at a press conference on Monday.
The team used data from general election polls and created a dynamic model that was heavily dependent on the political events that had occurred at the time of the analysis. The assumptions come from survey data and are intended to “capture the pulse of the people,” as well as take into account the two-party system, Putcha said.
Like with most elections, the team concluded that the presidential race was still quite fluid despite it being seven days away. With 5% to 6% of voters still undecided and some Republicans refusing to participate in polls, data could be distorted as a result, Putcha said.
Putcha added that history has shown that 75% of undecided voters choose the underdog, in this case Trump, as demonstrated through data collected by battleground states such as Arizona, Florida and North Carolina where the margins are getting smaller.
"The math is correct but the inputs from polls is what we worry about," Sloboda said.
The team cited their past predictions from 2008, 2012 and 2016, where the model had been correct during former President Barack Obama's administration but had fallen short in the last presidential election when Trump went head-to-head with Hillary Clinton.
Sloboda said that the team had done an analysis a week before the 2016 election and had accounted for Clinton's win but at a lower level. Putcha said that the model was still correct, and is still being used today, but the final prediction was imprecise.
"I'm sure we had the same confidence like that last time too, but we were wrong,” Putcha said.
In a forecast presented by Putcha, Biden would secure California with 60% of the popular votes, while only 33% of the votes will go to Trump, unsurprising for a historically Democrat state. On top of the popular vote, Biden would gain all 55 of California's electoral votes.
Unlike California, which has a clear winner, swing states, where the two major political parties have comparable levels of support among the population which makes it unclear.
Using Putcha's mathematical model, it is projected that in the swing state of Pennsylvania, Biden would gain 50.25% of the popular vote and Trump would gain 45%, with only a 5% difference in votes that could sway completely on election night.
Despite the uncertainties of shifting polls, Putcha assures that the prediction is correct with the mathematics, data collected from surveys and the assumptions made.