2020 vs. 2016: The comparisons we can make show us why Joe Biden is winning
One poll helps tell us the story of the 2020 election.
Welcome to Margin of Error, a newsletter from me about the polls and the way they are covered.
If you’re new here, thanks for signing up! If you find it interesting, I’d be grateful if you would encourage others to sign up.
If someone sent this your way or you found this post through Twitter or other channels, make sure to subscribe below.
Here’s where we stand with 20 days to Election Day:
National poll averages: Biden +10.6 (538); Biden +10 (RealClearPolitics)
Florida: Biden +4.6 (538); Biden +3.7 (RCP)
Pennsylvania: Biden +7.2 (538); Biden +7 (RCP)
Michigan: Biden +8 (538); Biden +7 (RCP)
Wisconsin: Biden +7.7 (538); Biden +6.3 (RCP)
Arizona: Biden +3.9 (538); Biden +2.7 (RCP)
FUN FACT: Kansas (Trump +6.6 in 538) is closer right now than PA/MI/WI.
And now, comparing 2020 to 2016—the good version.
Polling is an imperfect science. That’s even truer when you look at how we cover polls and how we try to neatly compare one election to another.
We compare the polls at this point to how they looked four years ago. We assume that they could be as wrong now as they were then, without acknowledging everything that’s changed in the four years since. We ascribe traditional narratives to issue-based polling that we think reflect how they will affect the race.
In the past week or so, a lot of people have even started making time-left-in-the-race comparisons. This time last cycle, for instance, was the point in the 2016 race when the “Access Hollywood” tape released, and could there be something like this in 2020?
In this edition of Margin of Error, I want to focus on the comparisons we can make between 2016 and 2020. That’s why the Pew Research Center’s latest survey provides a rare opportunity for a reliable, pound-for-pound historical comparison.
Its topline shows former Vice President Joe Biden leading President Donald Trump by 10 points, which is very much reflective of the current national average. But it’s also a valuable historical document, because it allows us to get perhaps the clearest glimpse of how the electorate has moved away from Trump over the past four years.
Some background: In 2018, Pew released a snapshot of the 2016 electorate. It was a validated-voter survey, meaning they validated respondents against actual voter rolls. It surveyed more than 3,000 people and found that their responses on vote preferences, 48-45 in favor of Hillary Clinton, closely mirrored the final results.
In its 2020 survey, Pew included responses from more than 10,000 registered voters and almost 12,000 adults. That’s a much larger number than most other firms’ surveys, and the Pew survey has a margin of error of only 1.5 points, though some subgroups are higher.
A few groups in particular help tell the story of Trump’s erosion.
Come on, man!
The caveat stands that this is one poll. But while many other public polls have shown that Trump is losing the support of women—an average of five polls released in the past week, for instance, shows him trailing by about 25 points—the Pew survey actually shows Trump’s biggest losses are coming among men.
Biden builds slightly on Clinton’s lead with women, leading in this poll by 16 points (to Clinton’s 15, as reported in Pew’s 2018 study). But Pew found that Trump won among men by 11 points in 2016. Now, according to Pew, Biden has a four-point advantage among men.
What kinds of men are shifting? White men, who now favor Trump by a much less significant margin (12 points) than the whopping 30 points by which he won that group in 2016. Black men have moved even more firmly in the Democratic column, favoring Biden 88-11 (a 10-point net improvement from 2016). One of the few bright spots for Trump is among Hispanic men, among whom he has narrowed the gap from 37 points to 24 points.
The only other direct comparison we have is among the subgroup of married men. The shift away from Trump there is similar to the overall erosion, from 30 points to 7 points. Indeed, a lot of the movement away from Trump has come among married people (from plus-16 Trump to just plus-3 Trump).
Culture war
The bread and butter of Trump’s base has long been white men without a college degree: In 2016, according to Pew, he won among this group by 50 points. The underrepresentation of this group in 2016 swing-state polls accounted for the polling error that missed his win and forced many pollsters to reconfigure how they weight their surveys.
Trump still leads with this group, but his lead has basically been cut in half: His support has dropped from 73 to 60, and Biden has improved on Clinton’s margin by more than 10 points.
There’s still a lot of introspection into media and polling failures in 2016, a lot of which has focused on these white men without college degrees. There’s been less examination during this campaign examining why this group has slipped slightly away from Trump.
Suburban war
One of the subplots of the 2020 election season has been the narrative of a suburban exodus from the president. This is reignited every time he fires off a tweet that plays on racist fears about suburban safety.
This poll shows the suburban exodus is real. Trump won suburban voters by 2 points, according to Pew’s 2018 accounting of his first win. Biden now leads these voters by 12 points. (In case you’re wondering, urban and rural splits have remained largely the same.)
Though there aren’t other direct comparisons, there are few other numbers worth noting. One of Trump’s more desperate-seeming focuses has been on suburban women —or, as he termed them in an oddly capitalized July 2020 tweet, “The Suburban Housewives of America.” The Pew poll shows that Biden is winning with suburban women by 19 points. Similarly, among suburban men, Biden is also leading, by 5 points.
Senioritis
Here’s another one for a future newsletter. (I’m committing myself to a lot in this edition!) There’s been a lot made about the senior switch toward Biden. This poll finds that to be true, albeit in smaller numbers than other surveys.
Some polls have shown Biden leading, even considerably, among the 65-plus group, among the president’s mismanagement of the virus. This poll shows the candidates dead even. (Pew found Trump won by 9 points among this group last time.) Biden also matches or expands on Clinton’s margins among other age groups—including, interestingly, the 50-64 age group, which Trump won by 6 last time.
Come together
There are still stories being written about an anonymous Bernie Sanders adviser’s grumblings about the state of the 2020 race. But one of the big stories of this campaign has been the unity among Democratic Party.
Part of that is that Democrats like their candidate much more this time: Biden actually enjoys a positive favorability rating (I’m planning to say more on this in a future newsletter!). In this poll, Clinton/Biden supporters view Biden more “warmly” than they did Clinton.
So it shouldn’t be surprising that Biden is winning 96% of Democrats, with a net advantage of four points over Clinton’s 2016 margin. The other story is how Biden is romping among self-identified independents. Trump won them by a point in 2016, while Biden leads by 18 points now.
He has also peeled off a bit of Trump’s 2016 Republican support, even as he appeals to them with a different message than Clinton. Where Clinton presented Trump as an aberration, Biden has consistently framed Trump, correctly, as the figure who shapes the Republican Party.
All of these elements are showing up, and have shown up, in the toplines of national and, especially, battleground state polls. The New York Times has done a good job of tracking vote shifts in interviewing more than 5,000 voters across the key states of Pennsylvania, Ohio, Michigan, Wisconsin, Minnesota, and Iowa. These voters self-reported the actual vote across all six states, and now they favor Biden by five points.
This is the key point:
“Mr. Biden leads with an overwhelming advantage among independent voters, who back him by 20 percentage points in both states.
And though Mr. Biden’s gains among white voters are broad, spanning both those with and without a college degree, he fares far better than Mr. Obama did among white college graduates, while faring worse among those without a four-year degree. As a result, Mr. Biden still trails narrowly in the precincts that flipped from Mr. Obama to Mr. Trump, while holding an overwhelming advantage in the smaller number of predominantly suburban precincts that backed Mitt Romney in 2012 and then supported Mrs. Clinton in 2016.”
In other words, to beat a dead horse: This isn’t 2016.
Thanks for reading Margin of Error. If you have any tips, comments, or insights about polling, email me at bplogiurato@gmail.com, or find me on Twitter @BrettLoGiurato.
If you liked what you read, share it with a friend and make sure to subscribe.