According to calculations by a prominent data scientist, Kamala Harris has a lead of 66 electoral votes over Donald Trump

Given Miller's stellar record in the 2020 cycle, it is extremely insightful to examine his prospects for this year's race. And as in the case of the Georgia runoffs, his view is shockingly contrarian: While most pundits and forecasters see a “dead heat” or a “tie,” Miller's numbers show that Harris-Walz is well ahead and the Democrats have a wide advantage , which settles into a remarkably stable pattern.

The Miller formula translates betting odds into votes

Miller's model is based on the “prices” posted on America's most active political betting site, PredictIT. It is a highly liquid market with an average of 37,000 shares traded per day. His principle: After excluding bets on third-party candidates, the odds of the two candidates reflect exactly their share of the popular vote. If a candidate's price on PredictIT is 54 cents, that 54 cents can be used to estimate vote shares that will be lower but still track those odds. By analyzing the results of every presidential election since 1960, Miller also concludes that popular vote shares reliably predict what percentage of the popular vote each candidate will receive. “The core of this model is 60 years of presidential election history,” admits Miller. He emphasizes his respect for the work of Allan Lichtman of American University, who also bases his well-received predictions on presidential election history.

“The surveys are based on much smaller samples than PredictIT,” says Miller. “Typically they only interview 500 to 1,500. There is no consistency among the respondents. The pollsters all use different groups of likely voters. They also use different methods, so there is no consistent way to analyze their data. In contrast, tens of thousands vote with their money through PredictIT. If they're right, they turn 57 cents into a dollar. It's a significant return on investment. Participants are motivated to stay in the market until the end of the election to earn the profit, so the group of investors setting the prices is unified.”

Another plus point is that PredictIT is open 24 hours a day. “So you can sell your shares at any time,” Miller continues. “When the Democrats are not doing well, you sell the Democratic stocks and buy Republican stocks. The investors always say, “I have an idea of ​​what will happen on November 5th and I want to make money from it.” Tens of thousands of participants express their opinion on the outcome through their bets.” The bettors gather in a political marketplace that is similar to a stock market where investors use their best information to predict future prices for stocks, bonds or commodities.

From now until Election Day, Miller will publish the projected electoral vote (EV) totals for the two candidates in real time; He updated the numbers on his website Virtual Tout shortly after the PredictIT pricing change. As of October 2nd, 11 a.m., Harris-Walz leads Trump-Vance 302 to 236, according to Miller, a significant lead of 66 electoral votes. The bad news for Republicans: Since the Harris-Trump debate three weeks ago, the Harris ticket has been well ahead, and the propensity of bettors to change their minds is low.

The VP debate only moved the markets slightly

By Miller's standards, the Oct. 1 vice presidential duel did relatively little to change the trajectory of a race that appears to have reached some sort of steady state. “There were no changes to the forecasts throughout the 90-minute debate. But at midnight on Oct. 2, there was a slight shift in favor of Republicans, suggesting that Vance's performance was slightly better than Walz's,” Miller said Assets. “In this case, only six votes went from Democrats to Republicans, so it was a minor change that showed a slight advantage over Vance.”

Miller recently made an important adjustment to his model. “I’ve found that the model doesn’t work well when the odds are high for a candidate,” he says. “This is because when people perceive the prospect of a candidate winning as extremely likely, they will only bet on that candidate, skewing the prices too much in their favor. At the extreme end, the original model distorted the estimate of vote shares. This has been corrected since September 22nd.”

Miller noted that the turning point of the campaign was Trump's appearance at the National Association of Black Journalists convention on July 31, when he claimed that Harris had misled voters about their race. Before this faux pas, Trump had around 290 electoral votes, 20 more than needed to win. The day after the NABJ event, Harris took the lead and has been in the lead ever since. Shortly before the debate on September 10th, Trump briefly almost equalized. Days earlier, Trump received a sentence thanks to a New York state judge's decision to delay sentencing in the Republican standard-bearer's hush money case until after the election. “As of September 10, Trump was at 261, just nine shy of a win, and Harris was at 277,” Miller says. “At the end of the day, Harris had added 35 and came to 312.” Miller also credits Taylor Swift's support, which he received shortly after leaving the debate stage, which gave Harris an added boost.

Harris' electoral vote surged over the next two weeks, reaching a high of 337 electoral votes on September 20.

Since Harris peaked after the debate, her lead has fallen again. Is this the beginning of a resurgence for Donald Trump?

Since Harris reached that peak, her voter count has fallen from nearly 340 to the current 302, and Trump's voter count has rebounded from just 201 to 236. But for Miller, this shift doesn't mean Trump is making a comeback. He notes that Harris' position over the past week, which spans the vice presidential debate, has been consistently at just over 300 electoral votes. Above all, he says, bettors' views on who will come out on top appear to be firmly established. And for Trump to win back the White House, a large portion of bettors will have to move into his camp.

“I don’t think Harris’ decline is significant,” Miller said. “Investors in prediction markets sometimes change their minds, but not many of them do.” He points out the low volumes at PredictIT. “The last time we saw an increase was at the time of the debate and the Swift endorsement,” he adds. “There was a slight uptick for Trump after the vice presidential debate, but the last major event that changed the odds was the presidential debate. Now most people don’t change their bets but keep their contracts.”

The poll numbers show that Harris is currently on track to win most swing states

According to basic electoral math, Miller says, as of today, Harris has a majority of the major swing states and appears a sure winner in Wisconsin, Minnesota, Pennsylvania and Michigan. He adds that Harris also had Georgia in her column when she reached 337 EVs on September 20 at the peak of her post-debate surge. Now, he says, the modest Trump surge has put the Peach State back in play, and Trump also has a good chance in North Carolina. But for now, the states considered most important since the race began remain out of his reach.

The world has become a lot more dangerous in recent days as the dockers' strike threatens to drive up food prices and the Middle East teeters on the brink of all-out war. Big events could still change the course of this election. But for now, Harris is in charge, Miller says. The oddsmakers get it, and as expected, the polls are way behind the curve. According to bettors, Trump squandered a nice lead by stumbling at NABJ and botched a budding comeback by faltering in the debate. He may need a turning point or a blunder from Harris to turn the race around. It could happen. But bettors doubt it will happen, and as Miller has shown, bettors are the best at cutting through the fog of polls and pundits and getting elections right.