In the final days before the 2020 presidential election, polls generally pointed to a clear victory for current President Joe Biden. But when the votes were counted, it turned out that the polls had overestimated him: Biden won, but by the skin of his teeth. That, of course, was similar to what happened in 2016, when former President Donald Trump significantly outperformed the polls in Michigan, Pennsylvania and Wisconsin and pulled off a surprise Electoral College victory despite relatively accurate national polls.

This raises two questions for 2024: first, what would happen if the polls were off again? And second, how likely is it that the polls will be off by as much as they were in 2016 or 2020?

Trump or Harris could win comfortably

In 2020, polls overestimated Biden’s margin over Trump by about 4 percentage points in competitive states. As of October 30 at 11:30 a.m. Eastern, the margin between Vice President Kamala Harris and Trump in the 538 polling averages is less than 4 points in seven states: the well-known septet of Arizona, Georgia, Michigan, Nevada, North Carolina, Pennsylvania and Wisconsin. That means that if the 2020 election mistake were repeated, Trump would win all seven swing states and 312 Electoral College votes.

Of course, if the polls are wrong, it won’t necessarily benefit Trump. The direction of the election’s error is impossible to predict in advance, and polls have often overestimated Republicans in the past. In a scenario in which the polls overestimate Trump’s margin in every state by four points, Harris would win all seven swing states and 319 electoral votes.

Note that neither produces particularly close results, at least in the Electoral College.

Some degree of polling error is normal

Both outcomes – and everything in between – will be very much on the table next week. But are these scenarios actually likely, or do they seem more like external possibilities? Well, that’s where the work we’re doing on our election prediction model can be useful. As of October 30 at 11:30 a.m. Eastern, our forecast gives Trump a 52 in 100 chance of winning the White House and Harris a nearly identical 48 in 100 chance. The model arrives at that probability by calculating how many Electoral College votes each candidate would win, given a certain amount of voting error in their favor, and then adding up how many times each candidate wins in these simulations. (More about that in our methodology.)

Based on how many polls have not been released in the past, our election model estimates that the average election error in competitive states this year will be 3.8 points at the margin.* This error is not uniform across states, for example states with different demographics tend to have varying levels of polling error – but in general, when pollsters overestimate a candidate, they tend to overestimate them across the board. In other words, the model expects a polling error roughly the size of 2020 — though not necessarily in the same direction as 2020. (In 50 percent of the model’s simulations, Trump beats his polls, and 50 percent of the time, so does Harris. )

This point is worth dwelling on. Because our average expectation is that there will be a fairly significant election error at least half of the time, there is actually a very small chance that the polls will be perfect and that the election will go exactly as the polls suggest. Let’s look at this against the largest lead either candidate has in the seven swing states: Trump’s current two-point lead in Arizona. Nationally, our model expects the polling error in both directions to be greater than 2 points in 62 percent of the cases. In other words, there is only a 1 in 3 chance that polls will be off by less than 2 points (which we would historically consider a small polling error).

Given that all seven key swing states are so close to each other, even small electoral errors in the same direction can have a major impact on who wins the election. According to our model’s simulations, there is a 60 in 100 chance that either candidate will win more than 300 Electoral College votes — which Harris could do by winning five of the seven swing states and Trump six of the seven. By modern standards, I think it’s fair to consider this a huge victory – given how closely divided the country is, it’s relatively unlikely that either candidate will win much more than this (even to get 320 electoral votes, Trump would have to win an electoral vote). state like Minnesota and Harris should win a state like Florida).

Of course, the likelihood of an eruption in any case depends heavily on the outcome of the popular vote. If Harris wins the national popular vote by three points, she is much more likely to win the states that will decide the Electoral College than if she loses the popular vote by three points. This is clearly shown in the graph below, which contains all simulations from our model and ranks them based on the popular vote outcome.

As you can see, Trump is favored to win the election even if he loses the popular vote by 1-2 points, which is what our national polling average currently suggests. And if national polls turn out to underestimate him, with Trump winning the popular vote by 1-2 points, he would be favored to win outright.

Meanwhile, our model assumes that Harris must win the popular vote by 2.1 points to be the favorite to win the election, because swing states are more Republican-leaning than the nation as a whole. And if she wins the popular vote by 4.5 points (Biden’s popular vote margin in 2020), she is favored to win on her own.

Opinion polls are inherently uncertain. This is why we model.

So far we’ve said little about the actual polls themselves. And there’s actually some reason to believe that the polls may be more accurate this year than in the past. While the share of polls conducted or sponsored by Republican organizations has increased — something we’ve written about — the overall share of partisan polls is lower than in previous years, and the average rating of 538 pollsters is higher in the 2024 polls than in previous years. . All else being equal, that should lead to better polling than in 2016 and 2020. We’ve also seen fewer polls of the companies that Democrats overrated the most in those years.

However, the news is not all good. Notably, pollsters continue to report that it’s difficult to reach voters in the first place, and it’s possible that Trump supporters are still less likely to respond to polls — even high-quality polls. This means that pollsters still (or perhaps even more!) rely on weighting and modeling to get good estimates of public opinion. But the decisions they make matter a lot, and there seem to be especially big differences between polls that try to use these techniques to balance their samples based on party or past votes, and polls that don’t.

And this is the big, fundamental problem with pre-election polls: we don’t know what the demographic and political makeup of the actual electorate will be, so pollsters are only making the best guesses they can. There are always errors associated with these guesses, and there always will be.

And that’s where election models like the 538 become really useful. The goal of creating election forecasting models is not to provide a hyper-accurate, laser-like predictive view of the election that removes all error from the polls. Rather, it is intended to give people a good understanding of how the polls could be wrong and what would happen if they were. By analyzing possible errors and uncertainties in the polls, these models help us approach the election with a clearer picture of how likely each party is to win (and to what extent).

As we enter the final week of this election, it’s a good time to remember that uncertainty is an inherent part of polls and elections. That’s especially true this year, with deadlocked races in the swing states. Given that the polls are imperfect, we expect them to be slightly off in either direction. And if the polls do end up going off, given the closeness of the election, there will be a fairly wide spread in the Electoral College results.

In other words, we can summarize the current state of the race as follows: although Trump and Harris have roughly equal chances of winning the election, the final margin will not necessarily be close. In fact, there’s a pretty good chance that won’t be the case.

Footnotes

*We simulate potential polling errors for future elections using a fat-tailed distribution – specifically a Student’s t-distribution with five degrees of freedom (a parameter that increases or decreases the likelihood of surprising “tail” events in our simulations). This 3.8 point error is the spread, or sigma, of that distribution – analogous to the standard deviation of a normal distribution. 538’s distributions are slightly wider than those used by other forecasting models. This is because our model takes into account the fact that election misses have increased over the past decade. Therefore, our model expects more bearing errors than if we were to assume a constant level of error over time, as most other forecasts do.