Pollsters Say There’s No Way of Knowing How Pollsters Get Things Wrong
If anyone has a clue how pollsters got it so wrong in the 2016 and 2020 elections, it isn’t the pollsters, who have yet to do any self-reflection.
Polls infamously predicted a decisive Hillary Clinton victory in 2016, with some models giving her a 98%+ chance of winning. In 2020, while the polls were right that Biden would be declared the winner of the election, the margin by which the polls were wrong was actually greater in 2020 than in 2016.
In 2016 the RealClearPolitics average had Hillary Clinton ahead by 3.2 points while she won the popular vote by 2.1. The 2020 cycle had Biden leading Trump by 7.2 points, who ended up winning the popular vote by a 4.4 points, a margin of error nearly triple that of 2016.
As I wrote at the time, “When it came to the national polls they did come close to estimating the popular vote in 2016, but state polls is where things broke down. State polls were off by roughly five percentage points in 2016,” and they failed tremendously again in 2020, particularly in key swing states. The average poll predicted Biden would carry Florida state by 0.9 points, while Trump won it by 3.4. The average poll had Bidden up 6.7 points in Wisconsin, which he carried by only 0.7 points. The accuracy of polling in swing states gives much more predictive power than national polls (as Democrat support could be outsized in states like California, which is irrelevant because of the electoral college).
To give credit where it’s due, Rasmussen Reports was the most accurate pollster in 2016, almost exactly predicting Hillary’s popular vote margin and a Trump electoral victory.
It’s clear that nothing was done between 2016-2020 to try to improve the accuracy of polling, and we shouldn’t expect that work to be done ever. According to the Daily Caller:
Presidential polls surveying November’s election were the least accurate in 40 years and state polls were the worst in the past 20, a report released Monday shows.
While President Joe Biden’s support was only overestimated by about a point, Trump’s was underestimated by over three.
While the report from the American Association for Public Opinion Research outlines how far off the polls were, it was unable to directly attribute what led to the inaccuracy.
“We could rule some things out, but it’s hard to prove beyond a certainty what happened,” said Josh Clinton, a Vanderbilt University professor and chair of the 19-member election task force,. “Based on what we know about polling, what we know about politics, we have some good prime suspects as to what may be going on.”
One of the main “prime suspects,” the report notes, is a lack of response to polls in the first place, echoing other pollsters who have tried to diagnose why they were so off. Other reports have attributed polls’ growing inaccuracy to Republicans’ unwillingness to participate in them, exacerbated by former President Donald Trump repeatedly labeling any surveys he did not like as “fake.”
In other words, be skeptical of any polls you see in 2024. Even the pollsters don’t think highly of them.
Matt Palumbo is the author of Dumb and Dumber: How Cuomo and de Blasio Ruined New York, Debunk This: Shattering Liberal Lies, and Spygate
Don’t miss The Dan Bongino Show