Presumptive Democratic nominee Joe Biden speaks at a campaign event in Wilmington, Delaware, on July 28, 2020. | Andrew Caballero-Reynolds/AFP/Getty Images

Trust in Biden to handle the coronavirus crisis seems to be playing a role in his advantage.

New CBS polls show presumptive Democratic presidential nominee Joe Biden 6 percentage points ahead of President Donald Trump among likely voters in both Wisconsin and Pennsylvania, crucial battleground states which Hillary Clinton lost by less than 1 point in the 2016 election.

There’s still plenty of time for things to change until the election, and battleground state polling should be taken with a grain of salt, but the results suggest Biden may have an advantage in the Rust Belt states that helped Trump secure his victory in 2016 — and that Trump’s botched response to the coronavirus crisis is playing a significant role in Biden’s ratings.

While the polls, which were conducted by YouGov on behalf of CBS between August 4 and 7, show Biden in the lead, it is important to note the former vice president’s lead is within both polls’ margin of error, meaning Trump could actually be polling a little better than Biden. In Pennsylvania, pollsters found Biden ahead of Trump with 49 percent support to the president’s 43 percent. That poll has a 3.7 percentage point margin of error, meaning Trump could have up to 46.7 percent support and Biden as little as 45.3 percent.

The Wisconsin poll — which had a margin of error of 3.8 percentage points — found Biden leading Trump 48 percent to 42 percent.

In both Wisconsin and Pennsylvania, independent voters were found to favor Biden. Clinton lost this group in both states to Trump; and Biden is also outperforming Clinton’s vote share among white voters with and without college degrees. It should be noted, however, that the two data points aren’t directly comparable — while polling data in this case comes from likely voters who may or may not actually go to the polls, 2016 voting data comes from voters who did, in fact, turn out.

Pollsters found views on the coronavirus pandemic as having a strong relationship to candidate preference — in fact, in its analysis of the survey data, CBS found that views on the pandemic are more strongly associated with voting than views on the economy.

“Those who say Wisconsin’s outbreak is a crisis are voting for Biden in even larger numbers than those who say the economy is very bad. The small group who think the outbreak is not much of a problem back Mr. Trump in larger numbers than voters who say the state’s economy is good,” the analysis says.

Public perception of the president’s pandemic response is highly polarizing — and polls in recent months have shown that the public finds it to be the most important issue facing the nation. CNN’s polling expert Harry Enten has argued this is bad news for Trump, since historical polling data suggests that “whoever is most trusted most on the non-economic issue is likely to win the election.”

State polling should be taken with a grain of salt

Polling in battleground states is important — particularly given the US’s presidential elections are determined by the electoral college, not a popular vote. But state polling also has significant limitations, and Biden’s consistent lead in them (including other states like Michigan, Florida, and North Carolina) should not be seen as a surefire sign of his victory in those states and the overall election.

Consider that a Marquette Law School poll in Wisconsin in late October 2016 had Clinton up by 6 percentage points — the same advantage Biden has in CBS’s Pennsylvania and Wisconsin polls — but Trump ultimately won the state by 0.7 points.

As Vox’s Li Zhou has explained, there are many reasons that a number of state polls were off the mark in 2016 compared to the final election results. Some of those have been corrected for during this election cycle — for instance, in the run-up to 2016, some polls overrepresented Clinton voters because they failed to weight for education, and that’s no longer the case. (The CBS poll is weighted for education.)

But there are still plenty of obstacles. Polling is always a snapshot of a specific time, and ultimately can’t give definitive insight into the likelihood that someone sharing their preference with a pollster will actually show up at the voting booth on Election Day, nor can it necessarily predict the patterns of late-breaking voters who decide on their candidate in the final days before election (something that played a crucial role in Trump’s victory).

Adding to the uncertainty is that the pandemic makes predictions based on polling especially difficult, as Zhou explains:

Specifically, the use of vote-by-mail due to the coronavirus pandemic makes predicting the composition of the electorate that much harder. It’s unclear how closely turnout will match up with prior years because of public health concerns about physical polling places and questions around the number of people who’ll use mail-in ballots instead.

“It’s difficult to do a turnout model because you’re not sure who’s going to turn out. That’s going to be even harder in an election that has extensive vote-by-mail,” says University of New Mexico political science professor Lonna Atkeson.

Bottom line: the polling is promising for Biden, but polls are not to be confused with perfect predictions of outcome.


Support Vox’s explanatory journalism

Every day at Vox, we aim to answer your most important questions and provide you, and our audience around the world, with information that has the power to save lives. Our mission has never been more vital than it is in this moment: to empower you through understanding. Vox’s work is reaching more people than ever, but our distinctive brand of explanatory journalism takes resources — particularly during a pandemic and an economic downturn. Your financial contribution will not constitute a donation, but it will enable our staff to continue to offer free articles, videos, and podcasts at the quality and volume that this moment requires. Please consider making a contribution to Vox today.