Are polls of the presidential race reliable this time?

Mike Bebernes
·Editor
·6 min read

“The 360” shows you diverse perspectives on the day’s top stories and debates.

What’s happening

Polls on the 2020 presidential race paint a clear picture: Joe Biden holds a significant lead, and Donald Trump’s chances of winning reelection are slim. Given the state of the campaign, which has been remarkably static for months, one might expect Democrats to be prepping their celebrations and Republicans to be resigned to the inevitable.

But with just over a week left until Election Day, Democrats are racked with worry about the outcome despite Biden’s lead in the polls.“Every time I get too happy I slap myself and stick my hand over a fire,” a Democratic lawmaker told Politico. The reason for their hesitancy, many say, is that they’ve been in this position before only to watch as the seemingly unimaginable slowly became reality.

Hillary Clinton was considered an overwhelming favorite in the final days of the 2016 race. She held a national lead of about 3 points and was ahead in polls of crucial “blue wall” states like Wisconsin, Michigan and Pennsylvania. But Trump pulled off narrow wins in each of those states en route to an Electoral College victory that shocked most political analysts.

The echoes of 2016 have partisans on both sides of the aisle asking a simple question: If polls got it so wrong then, how can anyone trust that they’re right this time around?

Why there’s debate

The errors of 2016 led to a period of introspection for many in the polling industry as they worked to find out what went wrong. Lessons learned during that process have resulted in polls that are more accurate, they say. The most common mistake in statistical models in 2016 was failing to adjust for education in their statistical models, an omission that made polls look more optimistic for Clinton. Current models account for education, making them more representative of the electorate, experts say.

Others say the 2020 campaign is much more predictable than the historically volatile contest four years ago. In 2016, undecided voters largely swung in Trump’s direction in the final days of the race, and turnout in rural areas of the country was higher than expected — both outcomes that caught pollsters by surprise. A much smaller slice of the electorate is undecided this year and turnout is expected to be high all over the country, which gives Trump less space to make up ground. Biden’s lead is also larger than Clinton’s was, meaning polls could be off by roughly the same amount they were last time and he could still win.

Polling experts repeatedly emphasize that their models are measures of what is likely to happen, not what will happen. Trump’s surprise win in 2016 was a result of the reality that sometimes unlikely events occur and the same thing could happen again, they say.

It’s also possible that some other confounding factor, like perhaps the unprecedented surge in mail-in voting, might later prove to be skewing the polls. Statistical models, experts like Nate Silver say, also have no way of accounting for any attempt to override the will of the voters, such as if GOP-controlled state legislatures attempt to send their own slate of electors to the Electoral College, a possibility that’s been floated by some analysts.

Perspectives

Pollsters have adjusted their models to avoid the errors of 2016

“There is always a chance of a systematic polling error, even when the reasons aren’t evident in advance. … Yet at the same time, many of the major causes of error in 2016 seem somewhat less acute.” — Nate Cohn, New York Times

Biden’s lead is large enough to withstand even a significant polling error

“If the polls are as wrong in 2020 as they were in 2016, Biden would win the presidency anyway.” — Philip Bump, Washington Post

Polls could miss again if GOP-favorable groups turnout in unexpected numbers

“Every election there is some new challenge for polling that’s not really anticipated or where there isn’t a solution in advance. The likely high turnout rate and the different patterns of turnout are the biggest challenges this year. We just don’t know.” — Political scientist Barry Burden to Chicago Tribune

Trump has a lot less room to improve than he did in 2016

“There were a lot more undecided or third party voters at this point in the 2016 cycle. ... undecided/third party group of voters were a pool that Trump could attract to make up the deficit he had to Clinton. … Given that there are far fewer undecided/third party voters this time, the chance of that happening again is slimmer than in 2016.” — Harry Enten, CNN

No polling model can account for an attempt to steal the election

“Our forecast assumes that the election is free and fair — at least to the extent that past elections that we used to train the model were free and fair. … Beyond that, it’s hard to estimate the probability that Trump could steal the election to any degree of precision.” — Nate Silver, FiveThirtyEight

The media has gotten better at communicating the uncertainty in the polls

“I always contended that the polls in 2016 were not nearly as far off as the interpretations of them were.” — Poynter Institute senior faculty Al Tompkins to PBS NewsHour

The 2016 race was much more vulnerable to a surprise outcome than 2020 is

“It's not 2016 anymore. So much has changed in four years: Trump is no longer the new disrupter, he's a president trailing multiple incendiary controversies as he seeks four more years. And Biden isn't Hillary Clinton: He carries far lower negatives and higher support in the Rust Belt states that put Trump in the White House.” — Richard Galant, CNN

There’s still time for the state of the race to shift in Trump’s favor

“Uncertainty remains primarily because even though Biden’s lead is big and has been remarkably stable, things could change and it could shrink. And if it does shrink, we’d see that a lot of things have not changed since 2016. It continues to be unclear if pollsters can more precisely gauge public opinion in the key Midwestern swing states, and the Electoral College has a large bias toward Republicans.” — Matthew Yglesias, Vox

Optimistic polls won’t make Democrats complacent like they were in 2016

“Anxiety is driving liberals to the polls in record numbers. ...In other words, fear of a 2016 replay may be the reason we won’t have one.” — Jonah Goldberg, Los Angeles Times

Is there a topic you’d like to see covered in “The 360”? Send your suggestions to the360@yahoonews.com.

Read more “360”s

Photo illustration: Yahoo News; photos: Getty Images