With control of the U.S. Senate on the line, a new tool is tracking the battle in near real time.
Throughout the day, the platform pulls data from the prediction markets Kalshi and Polymarket, and then uses statistical models to simulate a million versions of the midterm elections, with outcomes for each of the 35 races in play.
“Every 10 minutes, I’m running a gigantic simulation,” says Thomas Miller, the faculty director of the master’s in data science program at Northwestern University, who created the approach.
He calls it “NowCasting.” Prediction markets change very quickly, and so do the forecasts. “There’s value in that they’re very responsive to the news,” he says.
Some research suggests that prediction markets can be more accurate than polls. Still, the price that traders are betting on a candidate is not the same as a probability—if it costs 60 cents to bet that a Republican candidate will win, that’s not the same as a 60% chance of winning. For one thing, a large bet from a single trader could theoretically distort the price.
Platform fees and other pricing dynamics also affect the numbers. The prices of bets on all candidates in an election can add up to more than $1, even though their probabilities of winning should add up to 100%.
“I’m a data scientist . . . not a pollster”
Miller’s approach adjusts the prices and combines them with statistical modeling used in finance and econometrics. “It’s data driven—I’m a data scientist, not a politician, not a pollster,” he says. “I’m using the data, and using models that are trusted and that have been used for many years.”
Miller has been working with data from betting markets since 2020, when he correctly forecasted that Joe Biden would win using data from a market called PredictIt.
In January 2021, in the runoff election for Senate seats in Georgia, he correctly predicted that Jon Ossoff and Raphael Warnock would win. In 2024, when some other forecasters predicted that Kamala Harris would win the presidential election, he predicted that Donald Trump would win.
What’s the outlook for Tuesday?
At the point that we spoke on Friday, the latest model suggested that Republicans only had a 17.5% chance of keeping Senate control, down a couple of percentage points from earlier the same day, and a steep drop from the day before.
On Sunday, after surprise news that Michigan’s Democratic congresswoman Haley Stevens had endorsed the Republican candidate for Senate, Mike Rogers, over Abdul El-Sayed—Rogers’s chances ticked up slightly on the platform, but only temporarily, and nowhere near enough for a win.
By early Monday morning, with Election Day (Tuesday, November 3) just one day away, the platform predicted that Democrats had a roughly 90% chance of gaining control of the Senate.