Greensboro's African American Community Newspaper since 1967

Why the pundits and junk polls got the midterm elections wrong


I am angry at the chorus of armchair pundits who created a dominant media narrative around the red wave with little evidence of it. I’m furious about that because for those of us who do the work on the ground to persuade to vote, it was actually a challenging narrative environment when voters and all of us are being told, “This is going to be a red wave or a red tsunami,” as if, for instance, women voters had amnesia from the summer around the overturning of Roe v. Wade and were just focused on the economy and inflation. Not true, it turns out. Or Black voters, yet again, basically the conscience of America, turning out and showing up and really voting our values around racial justice and freedom and resilience.

– Dorian Warren, Co-president, Community Change and Community Change Action

A little over two weeks before Election Day, New York Times columnist David Brooks helpfully explained Why Republicans Are Surging.

The only problem: they weren’t. And they didn’t.

Brooks wasn’t alone. Fox News hosts Jesse Watters and Jeanine Pirro bet Geraldo Rivera $1,000 the GOP would win the Senate and the House. CNN’s Chris Cillizza offered up the following headlines: Why the midterms are going to be great for Donald Trump, Why Republican attacks on crime have been so devastating for Democrats, and The bottom is dropping out of the 2022 election for Democrats.

They weren’t, they haven’t, and it didn’t.

While a few House races remained too close to call, President Biden’s party has lost at least six seats, giving control of the chamber to Republicans. “However, Democrats flipped two Senate seats increasing its majority with a win by John Fetterman of Pennsylvania and Rafael Warnock recently winning the Georgia runoff held on Tuesday, Dec. 6.” It was hardly the 20- to 30-seat Republican gain in the House many forecasters predicted, and decidedly not “great” for Trump-endorsed candidates in competitive races, about 70 percent of whom lost.

It was also the first midterm election since at least 1934 that the President’s party hasn’t lost a state-legislative chamber; in fact, Democrats took complete control of three new state governments -- Michigan, Minnesota, and Vermont – and flipped the Maryland and Massachusetts governorships and the Pennsylvania state House.

Predicting a loss for the President’s party in a midterm election usually is a safe bet. The President’s party has lost seats in Congress in every election except two since World War II. The only exceptions have been 1998, when the President’s party gained five seats in the House and lost no seats in the Senate, and 2002, when the President’s party gained eight seats in the House and two seats in the Senate. Post-war, control of the House has flipped eight times and control of the Senate 10 times.

How did some of the most prominent voices in the media get it so wrong? One mistake is relying on outlier polls and unreliable polls, as David Brooks did. On October 20, the day his “Republicans are Surging” column appeared, an average of “generic ballot” polls showed Republicans with an advantage of just one tenth of one percentage point, and Democrats ahead in the four key Senate races. Brooks based his analysis on a single poll that found a four-point Republican advantage.

Closer to Election Day, however, even these polling averages shifted in favor of Republicans, thanks to what political strategist Simon Rosenberg called “a ferocious GOP campaign right now to flood the zone with their polls, game the averages, declare the election is tipping to them.” Political data specialist Tom Bonier noted that many of these polls assumed “an older, Whiter, more male electorate.”

Rosenberg told MSNBC’s Joy Reid, “This is an unprecedented massive campaign by the Republicans to game the polling average. And it’s disappointing to me this wasn’t caught earlier by many of the people that do this that are on TV and do this for a living.”

A bigger problem was this polling mirage served to confirm some pundits’ pre-existing biases, underestimating motivating factors like reproductive rights and the threat to democracy, that were not important to them personally, and overestimating the role of inflation and the false narrative of rising violent crime.

The American electorate is changing, growing more racially and ethnically diverse. Our pundit class – those whose opinion columns are published by major news organizations and who are given network and cable tv platforms to wax political – does not reflect this diversity. Until it does, it will continue to suffer from the blind spots that not only skewed predictions about the election, but potentially sabotaged it.

Marc H. Morial is the president and Chief Executive Officer of the National Urban League, the nation’s largest civil rights organization.