The State of the 2024 Election: Two Weeks Out

Ratings Changes: None this week.

Weekly Recap: Historical Polling Error and the Dobbs Effect

The word “tossup” seems to be the most common descriptor used by news outlets to summarize the 2024 election—so much so that it’s becoming a bit of a cliché. On the one hand, it is perfectly understandable for the media to cover the election as such: polling is close, Harris’s campaign is fretting over a perceived tightening of the race in Trump’s favor, and the vast majority of pundits seem equally uncertain as to who might have the upper hand. 

On the other hand, there seems to be a clear discounting of Harris’s obvious strengths in the election: her net personal favorability is a whole nine points higher than Trump’s; the Harris campaign’s fundraising numbers continue to blow Trump’s out of the water (between July 1 and September 30, Harris’s campaign committee raised $633 million, while Trump’s committees only raised a combined $340 million); and predictive indicators such as the Washington primary (discussed in an earlier article) paint a rosier picture for Harris than many polls do. 

Why might news outlets be less likely to emphasize these points? Part of it may be bad memories from past elections. 2016 is the most infamous example; almost every prominent news outlet, pundit, analyst, and—most importantly—pollster, was thoroughly humiliated after confidently assuming Hillary Clinton was a lock for the presidency. 2020 saw a similar phenomenon; after Joe Biden only narrowly defeated Donald Trump instead of securing the near-landslide many polls predicted, pollsters faced even more backlash. In 2024, then, many are operating under the assumption that the polls will be off in Trump’s favor again. History shows, however, that this should not be taken as a given. Furthermore, Harris may have another card up her sleeve: a phenomenon resulting from the overturn of Roe v. Wade known as the “Dobbs Effect.”

A Brief History of Recent Polling Errors

Polls are never accurate—that is, they rarely exactly correspond to the final results of the election, only approximating them at best. Additionally, when a nationwide election is taking place, there tends to be a polling error “favoring” a certain party, i.e. the polls show better results for a certain party than what pans out in the actual election. For example, as stated before, polling in 2016 and 2020 tended to be biased against Republicans and towards Democrats when compared to the real results. These two elections should not be seen as part of a universal pattern, however. Let’s compare the nationwide polling error for presidential and midterm elections between 2008 and 2022 (all numbers come from RealClearPolitics polling averages):

2008: D +0.3

2010: R +2.6

2012: R +3.2

2014: D +3.3

2016: D +1.1

2018: R +1.1

2020: D +2.7

2022: D +0.3

As one can see, there is essentially no pattern in whether or not the polls favor Democrats or Republicans—in fact, the polling errors in 2016 and 2020 were not especially statistically significant when compared to other elections. Indeed, when looking at 2016 in particular, it seems that Trump’s upset victory was less a function of polls failing to capture some secret base of support for his candidacy, and more that few analysts took the time to consider that a Trump victory was well within the margin of error—to most, such an outcome was too unimaginable to see as realistic. 

Even in the 2020 election, which had effects that pollsters are still reeling from, the miss does not stand out too greatly. Certainly, the picture is different on the statewide level: certain states that were expected to go handily for Biden, such as Wisconsin and Michigan, ended up going for him only narrowly. In 2022, however, many key Senate and gubernatorial races in which Republicans led in the polls went to Democrats instead—including in many 2024 swing states. Despite this, some pundits seem to believe there are only two possibilities in these states: that the polls will be spot-on, or that there will be yet another polling error missing Trump’s support. 

All this considered, it would be wrong to assume that simply because Trump is on the ballot, the actual results of the election will be more favorable to him than polling predicts. This is especially relevant because the main factors some have proposed to explain the seemingly consistent “shy Trump voter” phenomenon—voters being afraid to admit they support Trump, Trump voters having lower social trust and thus being less likely to respond to polls, and the COVID pandemic making it more likely for Democrats to respond to polls since they were more likely to be at home, among others—have much less relevance today. The pandemic is over, and Trump—a now-former president and domineering figure in American politics for nearly a decade—is a known entity whose followers confidently profess their support today. Do not be surprised, then, if, on Election Day, it is Harris who does better than what the polls predicted. 

The Dobbs Effect

Abortion, an issue at the forefront of the 2022 midterms, has featured surprisingly sparingly in analyses of which issues might tip the race. It is not necessarily the most important issue on voters’ minds, but it consistently remains among the topmost influential topics. A recent Gallup survey, for instance, shows the economy as the issue most commonly named as “extremely important” by voters, but abortion is still in the top ten. Where the issue of abortion really stands out, however, is in which candidate voters trust to handle it. While voters trust Trump over Harris by 9 points on the economy, they trust Harris by a whooping 16 points on abortion. And while 11% of Democrats stated Trump had an advantage on the economy, 16% of Republicans said Harris would handle abortion better. 

If Harris is able to take advantage of her stance on abortion, she would greatly benefit electorally. Indeed, in 2022, abortion was arguably the largest benefit to Democrats electorally—this benefit is sometimes called the “Dobbs Effect.” Before the Dobbs v. Jackson decision overturned Roe v. Wade in June of that year, Democrats were struggling to find a coherent message to voters. They faced what seemed like an imminent Republican wave that would take advantage of Biden’s low approval rating, a sluggish economy with high inflation, increasing crime rates, and a perceived cultural overreach by the Democratic Party in the wake of the pandemic and the protests over the death of George Floyd. Once Roe was overturned, however, Democrats suddenly had something to fight for: abortion rights across the country. It took a few weeks for the results to materialize, but Democrats soon surged in the polls, going from being down by two points to being up by one within the span of two months. On Election Day, Democrats performed remarkably well for an incumbent party in a midterm year, especially in swing races where voters listed abortion as one of the most pressing issues. 

The Dobbs Effect appears to still be present; a 2023 initiative to enshrine the right to abortion into right-leaning Ohio’s constitution, for instance, passed by a sizable 13 points, and the Supreme Court’s approval rating still remains low. And for as long as the precedent that Roe v. Wade established remains shattered, the political benefit to Democrats—or, more accurately, the political harm to Republicans—will remain. This is especially true among female voters and young voters, who helped drive Democrats to success in the 2022 midterms and may do so again this year.

The opinions expressed within this piece represent the views of the author alone and do not necessarily reflect the views of The Jefferson Independent.