And finally, “Pollsters remind us that there are only so many ways to predict the future,” Globe DC reporter Tal Kopan wrote earlier this month. “Ultimately, some voters will make their decision at the last minute.”

Here's what we know about the state of the presidential race.

Both national and state polls consistently show an incredibly close race and are not designed to achieve the precision needed to detect swings of one or two points, experts say. Keep in mind that there is often a margin of error of several percentage points. So if a poll shows a one-point lead for Harris, as a recent Suffolk University/USA Today poll found, that means a margin of error of 3 percentage points Harris could be either ahead or behind.

Therefore, survey averages were taken to identify trends over time. Averages are typically formed from a selection of high-quality publicly available surveys and take into account sample size, recency, and other factors. If we look at the national averages, the trend over the last few weeks shows a slight erosion of Harris' very narrow lead.

Averages calculated by FiveThirtyEight and the New York Times each show that Harris' national poll lead has shrunk by a point or two since mid-September, while the Cook Political Report shows that the race has remained essentially stable over the same period.

Looking at the polling averages of the battleground states compiled by FiveThirtyEight and the Times, the widest margin among the seven battleground states is just two percentage points, making the Electoral College a true tie. FiveThirtyEight shows Trump with a two-point lead in Arizona and Georgia. All other states show a lead of one point or less between Trump and Harris.

Courtney Kennedy, vice president of methods and innovation at the Pew Research Center, urges those concerned about the outcome to be clear about what polls can and cannot do.

“I'm often asked, 'Can I trust the polls?' And I push back and say, 'How can I trust the polls?' If you ask me, 'Can I trust the polls to reliably tell me who is going to win a close race?' The answer is no,” she said on Friday to the Globe. “They are not designed to do everything right when the race is within a few percentage points. If the question is, “Can I trust polls to tell me how Americans reacted to the Dobbs decision or how Americans feel about the economy?” Polls are absolutely up to the task.”

Polling averages allow us to look beyond some of the individual errors that might occur in a particular poll to see the bigger picture, Kennedy said. But they also have their limits.

“If the polls are wrong this year, there's a good chance they'll go in a similar direction, at least that's what we saw in 2016 and 2020. The polls actually had the same weakness. So we should not harbor false hope that averaging will miraculously fix the problem,” she said.

There is also evidence that polls are less accurate when Trump is on the ballot.

“They have absolutely struggled to fully measure his support base in the years he has been running,” Kennedy said, adding that a Pew analysis found that the majority of pollsters have changed their methodology since 2016 Address survey problems of previous cycles.

“But it's not like, 'Oh, we figured out the silver bullet.' Everyone has fixed the problem and we're good to go.' Instead, pollsters try a number of different corrections, and we hope this leads to greater accuracy overall.”

Ultimately, we will receive much less information about the state of the race. Large polling stations typically stop releasing new poll data in the days immediately before the election.

Christina Prignano can be reached at [email protected]. Follow her @cprignano.