The polls for the 2024 presidential election show a painfully tight race between Kamala Harris and Donald Trump, a race that could go either way, with little to divide the two candidates among the seven battleground states likely to determine the outcome.

But since 2016, when Republican candidate Trump pulled off a surprise victory over Democratic candidate Hillary Clinton, a result that was not widely or clearly expected in national polls, the polling industry has been closely scrutinized for its accuracy.

Similar issues in subsequent years, such as President Joe Biden’s overestimation of support for Trump in the 2020 presidential race, and underestimation of the size of the Democratic vote in some key midterm races in 2022, have kept the issue alive.

But polling companies have put a lot of work into revising and refining their models, trying to recalibrate their assumptions to better capture representative samples of American voters, thus restoring their statues as barometers of the nation’s thoughts and feelings .

And that’s an important point: polls should measure the temperature at a point in time, a snapshot rather than a hard-and-fast prediction of what will happen at a later time. Events happen, thoughts change, and so results differ from what polls implied.

On November 1, 538 – which takes a weighted average of polls and models the results to provide a probability-based prediction on who will prevail – the simulations show Trump winning 53 times out of 100 to Harris’ 47 . and-neck.

Yet perhaps this is all an illusion. There is a chance that the winner will emerge with a much stronger than expected performance, a scenario that some observers say is likely. And that would be an outcome predicted by few of the country’s pollsters.

With this in mind, Newsweek Ask US polling experts: Are we on track for another election miss in the 2024 election? And where are the polls most at risk of miscalculation for either candidate? This is what they said.

Charles Franklin, Director, Marquette Law School Poll

Next week we’ll find out whether the polls were generally right this year or not. Clearly, the polls and polling averages indicate a very close race between swing states and national popular votes. That contrasts with polls in 2016 and 2020 that showed larger Democratic margins. So this year the polls say we really don’t know who’s going to win because it’s so close. We shouldn’t be surprised if Harris wins, and we shouldn’t be surprised if Trump wins either. Both seem equally likely to win.

Mike Traugott, Research Professor Emeritus, Institute for Social Research, University of Michigan

Support for Kamala Harris and Donald Trump is all but ingrained, and there are very few people left to convince them. The outcome will therefore depend on the turnout and the ability of each party to get its supporters to the polls or get their hands on a ballot paper.

This means that the accuracy of the polls will depend on their turnout models – how they estimate likely voters.

Furthermore, I would expect a very large gender gap in support for the two candidates, perhaps enough to tip the outcome to Harris. For example, if we look at the national polls, most show the race to be very close, but the internal numbers show gender gaps of varying sizes.

This is troubling in a number of ways – women make up the majority of the electorate and are more likely to vote than men – and could explain any discrepancies between the final poll estimates (this Sunday and Monday) and the actual voting results.

John Zogby, senior partner, John Zogby Strategies

I strongly believe that polling in battleground states was solid in 2016. They captured the downward trajectory of Clinton’s support, enough to show that it could mean a loss.

If we only look at the polls from the day before and treat them as a “prediction,” the experts will fail, not the pollsters. I think the polls show this is a tie for now.

We should be able to determine by the weekend if the damn thing breaks somehow. Think of the polls as a series of still photos and not as forecasters.

Josh Clinton, co-director of the Vanderbilt Poll

Who knows. With equal probability, I think the polls will be better because it could be a repeat of 2020, and pollsters have reacted to the 2020 election miss to change the way they adjust their results (which is why we see so many correlations between the polls) ; that the polls will underestimate Trump even with the adjustments if new voters choose Trump and don’t participate in the polls (as in 2020), or perhaps support for Harris is understated as pollsters adjust to gain a lead from the Democrat and Harris take away the raw data because they think a Harris lead is a problem with who responds and not with what the electorate will be.

Christopher Wlezien, Hogg Professor of Government, University of Texas at Austin

When it comes to whether we’re on track for another election miss, it’s important to keep in mind that polls often get it wrong: The presidential bipartisan share at the national level averages 2 at the end of the campaign Dropped -3 points.

It’s no surprise that the numbers are getting bigger as we take a step back from the election. This is important to keep in mind when evaluating polls, especially when margins are small and performance is evaluated based on determining the right winner. That is, polls can come close to the final vote share and still point to the wrong winner.

And then there’s the Electoral College, which shines a spotlight on state polls, where the errors are typically greater than what we see nationally.

Now if you’re wondering if we’ll see above-average errors this time around, that’s of course possible, although I suspect pollsters will do better than in 2020 because they have an incentive to do so and have made several changes to that. end.

How much better I don’t know, because it is difficult to say what exactly they are doing and to what effect, taking into account that it is not easy to know who will cast votes even if there is significant early voting, and to to make them respond. to surveys.

We do not know that the decisions that pollsters make in making their estimates are (or could be) of great importance to the reported opinion polls.

Are they better at representing Trump voters who were seemingly missed in 2020? Have they, as some analysts have suggested, corrected too much? What about new voters? This all remains to be seen.

In the meantime, given history, don’t be surprised if a candidate underperforms or overperforms in the polls by a meaningful degree.

Courtney Kennedy, Vice President of Methods and Innovation, Pew Research Center

There are reasons for optimism about the accuracy of the polls this year, but there are also reasons for pessimism.

If there is an election miss, it will not be due to a lack of effort. My colleagues and I conducted a survey that found that most national pollsters (61 percent) have changed their methods since 2016. That is, they have changed the way they sample people, the way they interview them, or both.

Compared to 2016, polls today are more likely to offer people multiple ways to participate (for example, online or by phone). This can help reach a more representative group of people, because no one approach works for everyone.

Also, the 2022 midterm elections were generally accurate, despite a wave of partisan polling predicting a broad Republican victory. FiveThirtyEight found that “polls were more accurate in 2022 than in any cycle since at least 1998, with virtually no bias toward either party.”

But there are also reasons to be very cautious about the pre-election elections this year. There’s no denying that both times Trump ran for president, he underestimated his support in most polls.

The basic reason seems to be that Trump supporters are less likely to take polls than people who are otherwise similar to them (e.g., people of similar age, education, race, etc.). This pattern is not easily resolved by pollsters.

There are some things pollsters can do that we know will help (for example, giving people the option to take the poll offline, making sure the weighting protocol is adjusted to get the proportion of Republicans versus Democrats correct).

But these tactics may not be completely effective. We won’t know until the votes are counted whether the changes pollsters made were fully effective.

In recent cycles, the risk of election misses has been particularly high in states like Wisconsin. One reason is that some pollsters sample state voter files. In Wisconsin, the data added to the voter file — data that tells pollsters which voters are Republicans and which voters are Democrats — contains many errors.

A recent study found that “fewer than half of (registered voters) in Wisconsin who are perceived to have a particular partisanship identify themselves with the ascribed partisanship, and only a third of those who were thought in the voter file to they were Republicans. identified as Republicans during an interview.

In fact, nearly 50 percent of those thought to be Republican in the voter base report voting for President Biden — far higher than support among likely Republicans in other states. Somewhat similar patterns occur in Minnesota and Michigan.”

Demographically, it is particularly difficult for pollsters to get an accurate picture of young adults. One reason is simple: sample sizes among young adults are often small (e.g., 100 to 200 interviews), meaning the estimates will be quite noisy.

Additionally, younger adults are more likely to move and change their phone numbers, making it difficult to have up-to-date contact information.