This writer provides regular updates on presidential election forecasts based on forecasts by renowned data scientist Thomas Miller, a professor at Northwestern University. Miller asserts that his model, which is guided by political betting data, is much better at forecasting than following individual opinion polls.
Miller stresses that his approach is based on the results of more than 60 years of presidential elections.
I became convinced that Miller had a better system while following his framework and forecasts during the 2020 election cycle. He called the presidential race for Biden within 12 electoral votes, and in the Georgia Senate runoffs that followed, he assumed Democrats Raphael Warnock and Jon Ossoff would win when polls were still They appear to trail their GOP opponents by a wide margin.
Regarding the Harris-Trump race, I published three stories between September 18 and October 3, and in each, Miller's methodology showed Harris with a wide lead, while polls showed a tight contest, and the media almost always described the contest as very close. From contact. In fact, the first article said that, by Miller's measures, the vice president could be headed toward a landslide victory approaching President Lyndon Johnson's landslide victory in 1964. And even in the October 3 update, Harris — as of the day before — was. The former president leads by a large margin of 66 electoral votes, 302 to 236.
Last week, the contest took another surprising turn, this time in Trump's favor
In a battle that already feels like the craziest ride on an undulating rollercoaster, Donald Trump made a powerful comeback over the week until midday on October 9. For Miller, this came as a surprise, because in the 20 or so days following the presidential debate, the race appeared to have entered a plateau with Harris consistently in the lead. But as of midnight on October 8, Miller's system gives Trump a slim advantage, 270 votes to Harris' 268. Miller said: “This is the first time in history that the race has gone from a seesaw to a landslide for one side, then to a seesaw, then to a quasi-landslide for the other side, and then back to a seesaw.” luck. “The dynamic of this election is that after one candidate comes out on top, the race always comes back to a tie. And that's where you stand right now. I'm following this minute by minute, and the results keep switching back and forth around the 270 electoral votes needed to win.”
It is important to briefly review Miller's model, which worked so well in 2020. It is governed by two governing principles. The first is based on the odds expressed by bettors that they will reap big wins if the candidate they are betting on, rather than the candidate they plan to vote for, wins on November 5th. To get these odds, Miller uses prices published on the most established political betting site, PredictIT, which trades an average of 35,000 shares a day bought or sold on both candidates. Miller eliminates the votes of third-party contenders, making Trump and Harris' prices equal 100% of the total.
He then publishes his methodology for translating probabilities into popular vote shares for each candidate. This task requires a number of modifications. For example, a 50% stake in PredictIT for Trump equates to less than 50% of the potential popular vote for the GOP ticket, in part because bettors are primarily men, often regular sports gamblers, and lean toward the Republican Party. On the other hand, the GOP ticket could win less than half the popular vote, another factor Miller calculated in his analysis of the numbers.
Guideline Two: Projected popular vote percentages closely track the share of electoral votes each candidate receives. That's the data scientist's takeaway from studying every election since 1960. In the past few days, PredictIT's odds have narrowed from favoring Harris by several points to an actual tie, accounting for an electoral lead that now swings from one candidate to the next. Within narrow ranges of a few electoral votes (EVs).
Following a course of care is an exercise in electioneering
On his home page, The Virtual Tout, Miller posts a graph showing Democrats' share of EVs at the end of each day, overlaid with major events that appear to have swayed voters. The chart looks like a giant letter “S” lying on its side. In early June, the showdown between Biden and Trump looked like an even contest. That was my first “careless” moment. Following the June 27 debate, Trump surged ahead, and by the Republican convention in mid-July, Trump had more than 300 electric cars. This was Trump's first and only “landslide” phase. The Republican standard-bearer maintained a double-digit lead in EVs until his disastrous appearance before the National Association of Black Journalists on July 31, where he falsely claimed that Harris misled voters about her race.
On that day, Democrats took the lead for the first time. But Trump quickly recovered, and had almost equaled the number by August 5. This is the “second throw” point. The next day, Harris secured the nomination and appointed Tim Walz as her running mate. Biden's withdrawal gave Democrats a huge boost, pushing Harris up; It reached a peak of just over 350 electoral votes on August 11. The feat marked the culmination of the “second landslide” period, and this time Harris was ranked as the major victor just a month after Trump emerged as an easy winner.
Once again, Trump closed the gap, coming within seven EVs of Harris on September 9, the day before the debate. This is the “third throw” point. Trump's disappointing performance on stage in Philadelphia famously sent a slew of EVs into Harris' column. This time, the uptrend peaked on September 20 at 337 votes for Harris to 201 votes for Trump. The chart veered into “third landslide” territory, marking the second time in six weeks that Harris appeared headed for the exit.
In late September, the gap gradually narrowed. Miller thinks J.D. Vance performed slightly better than Walz in their October 1 debate. But Vance's good performance, he says, did almost nothing to improve Trump's chances. Two days later, Harris still held a huge lead of 58 EVs at 298 to 240. Trump's big surge began on October 4. In the four days that followed, he jumped to 270 points, while Harris fell by 30 points. Electric vehicles to 268.
Miller sees a stalemate until November 5, not continued movement toward Trump
I asked Miller if he saw a wave rising in Trump's favor, given the momentum he's shown in the past few days. “I don’t think so,” he answers. “Trump is still playing to his base. I haven't seen him change his rhetoric to reach more voters like a regular politician. So what explains the sudden jump in GOP numbers? He adds: “This may be due to fears of a large-scale war breaking out in the Middle East, and of American forces joining the conflict.” “The Biden regime has a traditional foreign policy of supporting allies militarily, leading to the view that could happen. Trump is more isolationist, and that attracts a lot of people who think about Afghanistan and Iraq and say: Why are we trying to do that? Why get involved in foreign wars? “A lot of people from all walks of life are not interested in us going to war.”
The pattern of huge increases that continue to shrink, repeatedly returning the race to an even balance, is emblematic of America's current predicament. “We are an incredibly divided nation,” Miller says. “The most likely trend over the next 26 days until November 5 is that expectations continue to return to volatility. Next month is going to be a crazy time.”
“Extreme travel lover. Bacon fanatic. Troublemaker. Introvert. Passionate music fanatic.”
More Stories
The FTC is moving to eliminate the “frustration” of canceling subscriptions for consumers
UK inflation September 2024
S&P 500 could hit 8,000 by the end of the decade: Ed Yardeni