Top data scientist sees Republican Senate chances rise slightly, but model still predicts that Democrats will flip 5 seats

This writer has been carefully following the Senate race forecasts on Virtualtout.io, a predictions site developed by Thomas Miller, a data scientist and professor at Northwestern University. The Miller approach first captured my attention when in January of 2021, his model posited that contrary to the polling, Democrats would win both of the two Senate races in Georgia. Miller scored again by calling for a Trump triumph in 2024.

Miller’s template for the current Senate contests once again starts with the “prices” posted on betting sites. He relies on the two platforms that host the most trading, Kalshi and Polymarket. Miller calibrates the prices from both sources so that they reflect percentage probabilities of victory or defeat in each of the 35 races.

He then uses multiple time series analysis that adjusts for “underlying factors” that tend to drive all the election “markets.” The overriding force by far, he says, is what Donald Trump just did or said. “The states don’t move independently of one another,” he told me. “If Trump does something voters don’t like, you’ll see the Republican numbers drop in all the states.”

On Virtualtout.io, a live “ticker tape” for all 35 races runs across the top of the screen. It shows that odds of a Democratic or Republican win that’s updated every 10 minutes, and also shows the change over the past 24 hours. Naturally, his data is far more current than the polling, which often reflects voters’ preferences several days or more before the day they’re posted.

Keep in mind that Miller’s numbers are a snapshot of today’s probabilities, and not a prediction of what will happen on November 3. But his data does show the Democrats in a strong position, as of now, to flip seats in five states, an outcome that would radically change the count from today’s 53-47 in favor of the GOP to 52-48 for the Democrats.

Virtualtout went live on October 23, but Miller collected numbers starting on September 17. On that date, his models showed a 34.8% chance the Republicans would keep control. As of early morning, October 26, that number had dropped to 23.3%.

The Republican side, however, is doing a bit better; its odds have risen 2 points from an “all time” low of 21.3% on October 24. It’s interesting to examine the interplay among the races, which show somewhat better chances for the Republicans than previous numbers in three states (Texas, Maine and also Michigan, a seat the Republicans had hoped to flip, but still favors Democratic candidate Abdul El-Sayed).

Let’s start where the GOP chances look remote and haven’t brightened. North Carolina’s dead: Miller’s model gives the Republicans a 4.5% chance of winning. Ohio was a near tossup six weeks ago. Now, incumbent Jon Husted’s got only a 32.5% shot versus Sherrod Brown. Alaska’s also looking dim. The GOP’s chances opened Miller’s derby at 33% and now stand at 24.8%.

Texas also started as near even before the Republican odds began to fall. But in recent days they’ve improved, rising from 34.2% on October 24 to 36.4% on October 26. The outlook in Maine’s also improved a bit for the Republicans, going from 32% for Republican Susan Collins on September 17 to today’s 37.9%.

And in Michigan, the GOP’s odds dropped to a low of 26.4% on October 5, but have risen to nearly 34%.

Miller’s numbers are highly fluid, and the Republicans in his model still harbor roughly one-in-three or better chances of winning in four states. So it’s by no means a done deal for the Dems. The slightly higher numbers for Republicans could mark a trend. An election that looked lacking in suspense could show plenty in the closing days.

This story was originally featured on Fortune.com