September 1–30, 2024 ·

We have been predicting a strong Democratic victory in the Electoral College since September 10, the day of the Trump-Harris debate. These results flow from a revised model first introduced with end-of-day forecasts from September 22, 2024.

As late as September 13, The Economist posted this headline: “Kamala Harris and Donald Trump are neck and neck.” Most political pollsters and pundits suggest that this will be a close election.

What makes The Virtual Tout® so different? Our forecasting models rely on prediction markets, not polls.

Political polls are snapshots of the recent past. Polls have limited predictive power when conducted many days before an election.

Prediction markets, on the other hand, are forward-looking. Investors anticipate what will happen on election day and place their bets accordingly. Just as the stock market is a leading indicator of what will happen with the economy, a political prediction market is a leading indicator of what will happen with an election.

The Virtual Tout® election forecasts are data-driven, informed by the history of US presidential elections. We describe forecasting models and methods in Top-Down Forecasting.

Consider current and campaign events in September.

On September 5 and 6, we learned that Liz Cheney and her father Dick Cheney endorsed Kamala Harris for president, joining many other Republicans in warning about the dangers of Donald Trump to American democracy and the world order. Also on September 6, we learned that Trump’s sentencing in the New York “hush money” case would be delayed until after the November election, which could explain the drop in forecasted Electoral College votes for the Democratic ticket between September 5 and September 8. Trump’s legal affairs can affect the presidential election.

On September 9, we were predicting a close election, with the Harris/Walz ticket commanding 277 Electoral College votes, only 7 more than the 270 needed to win.

Both parties’ election prospects changed dramatically on September 10, when we saw the combined effects of two critical events: the Trump-Harris presidential debate (with an estimated 67.1 million viewers) and Taylor Swift’s endorsement of Kamala Harris. On that day, there was a substantial increase in forecasted Democratic Electoral College votes.

In the Trump-Biden debate on June 27, Biden’s performance revealed the effects of his age. In the Trump-Harris debate on September 10, Trump revealed his personality. Most regard Kamala Harris as the clear winner of the Trump-Harris debate.

On September 12, the front page of The New York Times summarized campaign messaging from the presidential candidates:

To Trump, U.S. Is Failing.
To Harris, There’s Hope.

Also relevant to the political fortunes of the campaigns, The Wall Street Journal on September 12 reported a 2.5% August inflation rate with the associated headline:

Inflation Slows to 3-Year Low

On September 15, Secret Service agents prevented an assassination attempt on Donald Trump. This was the second assassination attempt on Trump.

On September 16 and on the days following September 16, many poll results were released, indicating higher voter preferences for the Harris/Walz ticket.

On September 17, American singer and songwriter, Billie Eilish and her brother Finneas endorsed the Harris/Walz ticket. Libertarian-leaning podcaster Joe Rogan praised Harris’s debate performance.

On September 18, the Federal Reserve cut interest rates by a half-point, the first cut since the beginning of the pandemic. This lowers borrowing costs for households and businesses, and could promote economic growth in the fourth quarter of 2024.

Current and campaign events in September inform the prediction markets. And The Virtual Tout® converts prices in prediction markets into Electoral College vote forecasts for the Democratic and Republican tickets (270 votes to win):

September 1: Harris/Walz (303), Trump/Vance (235)
September 2: Harris/Walz (303), Trump/Vance (235)
September 3: Harris/Walz (298), Trump/Vance (240)
September 4: Harris/Walz (298), Trump/Vance (240)
September 5: Harris/Walz (298), Trump/Vance (240)
September 6: Harris/Walz (288), Trump/Vance (250)
September 7: Harris/Walz (283), Trump/Vance (255)
September 8: Harris/Walz (278), Trump/Vance (260)
September 9: Harris/Walz (277), Trump/Vance (261)
September 10: Harris/Walz (312), Trump/Vance (226) Trump-Harris Debate Day
September 11: Harris/Walz (318), Trump/Vance (220)
September 12: Harris/Walz (317), Trump/Vance (221)
September 13: Harris/Walz (317), Trump/Vance (221)
September 14: Harris/Walz (316), Trump/Vance (222)
September 15: Harris/Walz (313), Trump/Vance (225)
September 16: Harris/Walz (323), Trump/Vance (215)
September 17: Harris/Walz (327), Trump/Vance (211)
September 18: Harris/Walz (333), Trump/Vance (205)
September 19: Harris/Walz (331), Trump/Vance (207)
September 20: Harris/Walz (337), Trump/Vance (201)
September 21: Harris/Walz (331), Trump/Vance (207)
September 22: Harris/Walz (327), Trump/Vance (211)
September 23: Harris/Walz (327), Trump/Vance (211)
September 24: Harris/Walz (317), Trump/Vance (221)
September 25: Harris/Walz (317), Trump/Vance (221)
September 26: Harris/Walz (312), Trump/Vance (226)
September 27: Harris/Walz (302), Trump/Vance (236)
September 28: Harris/Walz (302), Trump/Vance (236)
September 29: Harris/Walz (312), Trump/Vance (226)
September 30: Harris/Walz (308), Trump/Vance (230)

Trading volumes have been low since the September 10 Trump-Harris presidential debate. September 10 was a key day in the election, with the debate and the Taylor Swift endorsement of the Harris-Walz ticket. A recording of the Trump-Harris debate is available from ABC News:

A thoroughgoing analysis of the debate was provided by political commentators at MSNBC.

Over the past few months, there have been days (in addition to September 9) when we foresaw a close election. Predicted Democratic votes were very low in mid July—it looked like a Republican landslide. Since the September 10 debate, however, forecasts have been moving strongly in the direction of the Democratic ticket.

Time series figures such as this show massive changes in forecasted Electoral College votes in the 2024 presidential election:

Electoral College vote forecasts rely on closing prices in the PredictIt market/contract for the party that will win the 2024 presidential election. The average daily two-party trading volume in this highly-active market has been around 37 thousand shares.

Campaign events help to explain a strong Democratic win in the Electoral College. Note these dates with high trading volumes and associated increases in forecasted Democratic electoral votes:

July 21–23 (384 thousand shares over three days). Biden withdraws from the race and endorses Harris.
July 31 (102 thousand shares). Trump’s appearance at the National Association of Black Journalists. Trump falsely suggested that Harris misled voters about her race. July 31 represented a turning-point in the election, as forecasts moved from an anticipated Republican win to a Democratic win.
August 6–7 (188 thousand shares over two days). Harris secures the Democratic Party nomination for president and selects Walz as her running mate.
September 10 (158 thousand shares). Trump-Harris debate and Taylor Swift’s endorsement of Harris. Election forecasts suggest that the race may be moving from Democratic-leaning to a strong Democratic win.

Could current events and campaign activities affect future forecasts? Of course they could. We will keep an eye on events and associated prediction market prices as we generate and publish daily election forecasts until election day, November 5, 2024.

News releases from Northwestern University and the School of Professional Studies review our election forecasting methods.

Our methods have been recognized in the news: see Media.

The Virtual Tout® listens to all sides of the political debate. We post research results, election forecasts, and political commentary on this site.

Return to the log of News postings.

See the most recent news on the home page.