The Election Polls Are Wrong

Polls show Joe Biden ahead nationally and in key states, but do we believe it? Jonathan Jakubowski and Christos Makridis are among those saying no: "We predict that President Trump is going to win the 2020 presidential election—and win big," they write at the Hill. They argue that Trump supporters are less likely to admit their political allegiance to pollsters because it isn't "the socially desirable answer" (and there is evidence of that). They add that many polls fail to secure a representative pool of likely voters, or reach a minimum sample size of 1,000 voters. And the recent news about Hunter Biden's emails (questionable as it is) "may continue putting wind in Trump's sails," they write. "We don't pretend to have all the answers, but we do have some of the right questions. How are people actually feeling?"

• In that vein, Shawn McCreesh is scratching his head: He asks in the New York Times why it feels "so different on the ground" in Pennsylvania when polls show Biden winning "crucial" counties. Take Bucks County, north of Philadelphia: "This place—the land of hoagies and Bradley Cooper and Rocky Balboa worship and Tina Fey's 'Cousin Karen' accent—has transmogrified into Trumplandia," he writes. But he doesn't know whether it's a "fortissimo minority circling the drain" or "a blinking red warning sign that the polls" are wrong.
• Over at NBC News, David Wasserman analyzes 2016 and 2018 midterm polling and finds that "state polls chronically underestimated Republicans' strength in the Midwest and Florida." In the Great Lakes and Midwest states, polls underrated the GOP's final margin 28 of 33 times. Polls also underestimated the Republican lead in Florida in 2018 and North Carolina in 2016. But in the Southwest, Democrats' final margin was undershot 17 of 19 times over the two elections. "In the end, the only certainty in the polling world is some degree of error," he writes.

(More Election 2020 stories.)