Election forecast tracks a tight presidential race
Model updates election forecasts daily, driven by the betting market
For the second U.S. presidential election in a row, a Northwestern University data scientist is running a novel forecasting platform that updates the odds of a win by former President Donald Trump or Vice President Kamala Harris each day.
With this level of precision, followers can see how single events — such as a debate, campaign activities or legal rulings — might affect the potential outcome of the U.S. presidential election.
"A betting market isn’t asking people to give an opinion ... but to put their money down."
Thomas Miller, Faculty Director SPS Data Science Programs
“Tomorrow’s debate is critical,” said Northwestern’s Thomas Miller, who developed the platform. “Trump’s legal events also are critical. There have been massive changes in forecasted Electoral College votes in recent months associated with current events and campaign activities.”
Miller is the faculty director of the master’s in data science program at Northwestern’s School of Professional Studies.
Viewers can follow his daily predictions and accompanying analysis on his site, The Virtual Tout.
This platform follows the money
Miller’s system uses data from PredictIt, prediction markets in which users bet real money on political races. He then uses pricing data as input to his forecast for how the Electoral College will vote.
Using his daily forecasts, Miller can gauge responses to singular news or campaign events. When Trump received a sentencing delay for his New York “hush money” conviction, for example, Harris’s campaign experienced a predicted drop of 68 forecasted electoral votes, according to Miller’s model.
As of the time of this article, Miller’s system predicts the Harris-Walz ticket will win the November election with 289 electoral votes. Candidates need 270 votes to win the presidency.
Miller was nearly perfect in 2020
Miller’s models proved their uncanny accuracy during the 2020 presidential election — only predicting one state (Georgia) incorrectly. Miller has since learned from this error, uncovering innate biases that led to a Democratic Electoral College vote prediction that was 12 votes lower than the actual Electoral College vote.
“In 2020, I was wrong on Georgia because I did not account for the Republican bias in prediction markets,” Miller said. “I have the data from the 2020 election to gauge the degree of bias in prediction markets and to correct for that bias, if necessary.”
Regardless of the incorrect prediction for Georgia, Miller’s 2020 model was still more accurate than Nate Silver’s FiveThirtyEight. While FiveThirtyEight predicted 348 electoral votes for Biden-Harris, Miller predicted 294 electoral votes for the Democratic ticket. Ultimately, the Electoral College cast 306 votes for Biden-Harris.
Markets are better than polls for a few reasons
Miller says popular election forecasting systems, including FiveThirtyEight and The Economist’s predictions project, are inherently flawed because they use data from opinion polls. According to Miller, these data are old, compared to fast-moving news cycles.
Miller also notes another key advantage of prediction markets over political polls: Large groups of investors who stay in the markets until Election Day. These groups grow larger as Election Day approaches. Miller relies on prediction markets that have tens of thousands of investors, with thousands of shares traded each day. Typical opinion polls involve between one and two thousand respondents, with new respondents recruited for each poll.
“Prediction markets are more reliable than pollsters and pundits,” Miller said. “A betting market isn’t asking people to give an opinion or preference but to put their money down. When you put your money down, you believe what you are betting on. You might not like the outcome, but you believe it will happen.”
— By Amanda Morris
"Predicting the 2020 Presidential Election" Data Science Webinar
November 5, 2020 -- In this webinar, Data science faculty director Thomas W. Miller explains how The Virtual Tout® model works. Miller also reviews results from the 2020 presidential election.
Thomas W. Miller is faculty director of the data science program in the School of Professional Studies at Northwestern University. Miller developed The Virtual Tout, a model driven by prediction market prices, providing forecasts of victory in the Electoral College.
Northwestern University School of Professional Studies offers many degree and certificate programs, with evening and online options available. To learn more about how Northwestern University's Master’s in Data Science prepares graduates to advance their careers in data science, fill out the form below, and we will be in touch with you soon.