Expert Commentary

Election Beat 2022: ‘Known unknowns’ in the US midterms

We know reapportionment, redistricting, and a host of new laws affecting ballot access may give Republicans an advantage, writes media scholar Tom Patterson. The unknowns lie in the impact that each of those factors will have.

Known unknowns
(Gerd Altmann from Pixabay)

The midterm elections are less than a month away without much accounting by reporters for the “known unknowns” that could determine which party controls the House and Senate. Popularized by Defense Secretary Donald Rumsfeld, the phrase “known unknowns” refers to factors we recognize but don’t comprehend in contrast with unknown knowns, which are things that you think you know but turn out to be different.

In the context of 2022 midterms, the “knowns” are reapportionment, redistricting, and a host of new laws affecting ballot access. The “unknowns” are the impact that each will have.

We know a few things that provide a hint at their impact. As a result of reapportionment from the 2020 Census, seven House seats shifted between states. Republican-controlled states acquired five of them. Texas picked up two seats while Florida, Montana, and North Carolina each picked up one. Based simply on those numbers, Republicans should have a net gain of three seats from the shift. But it could only be two seats in that three of the states that lost seats have a Republican controlled legislature – Ohio, Pennsylvania, and West Virginia.

The likely net effect of redistricting is murkier. Republicans had more control over redistricting than Democrats. They control more state legislatures than do Democrats. Moreover, the states where redistricting is entrusted to an independent or bipartisan commission rather than the state legislature are more likely to be held by Democrats.

By one standard, however, Republicans do not appear to have gained all that much from redistricting. According to a recent analysis by the Brennan Center for Justice, 208 of the newly drawn 435 House districts would have been carried by Donald Trump in 2020 and 227 by Joe Biden.

Nevertheless, the presidential vote is not a reliable indicator of the likely outcome in competitive House districts. One reason is found in the “surge and decline” theory of midterm elections. Presidential elections bring out a “surge” of voters who, compared with those who also vote in the midterm, are more responsive to the mood of moment and disproportionately back the winning presidential candidate and the congressional candidates of that party. In the midterm election, these voters stay home (“decline”), which erases the in-party’s advantage. It’s a reason why the in-party has picked up House seats in only three of the 25 midterm elections of the past century.

The referendum theory of midterm elections also suggests that the 2020 vote is not a reliable basis on which to estimate the effect of redistricting. This theory holds that, after they take office, presidents make decisions that usually make more enemies than friends, leading to declining popularity that weakens the appeal of the president’s party in the midterm. The predictive quality of the referendum theory is apparent in the three exceptions to the in-party loss of House seats in the past century. In all three cases (Franklin D. Roosevelt in 1934, William J. Clinton in 1998, and George W. Bush in 2002), the president’s approval rating was above 60% at the time of the midterms. When a president’s rating has been near 40%, as Biden’s is, the in-party has always suffered a double-digit loss of House seats.

All this suggests, coming into the midterms, that Democrats had only a slim chance of holding onto the House and that reapportionment and redistricting have further reduced it.

The final “unknown known” is the impact of the more restrictive ballot access laws that Republican state legislatures have enacted since the 2020 election. The changes have put limits on early and mail-in voting. Exit polls in 2020 indicated that more than 60% of early and mail-in voters cast their ballots for Biden while more than 60% of those who voted on Election Day cast a ballot for Trump. Some of the new laws were more precisely targeted to reduce the Democratic vote. In last week’s Election Beat column, for example, I cited the Georgia law that restricts the number of Sundays on which early voting and registration can occur. Sunday has traditionally been a time when Black congregations after church services collectively register and vote.

Will these new laws reduce the Democratic vote by enough to tip some races to the Republican candidate? We won’t know until after the November election whether they will have made difference. But they could. Several of the most competitive House and Senate races are in states where Republican lawmakers after the 2020 election enacted restrictive voting laws.

If all the unknown knowns point toward a Republican victory in the midterms, they are not the only factors in play. The Supreme Court’s abortion ruling, for example, has activated Democratic voters. Typically, the “enthusiasm gap” for midterms favors the party out of power, as it did earlier this year. In March, Republicans were 17 percentage points more likely than Democrats to say they were very interested in voting in the midterms. In the most recent Politico/Morning Consult poll, Democrats were as likely as Republicans to express strong interest.

Thomas E. Patterson, Bradlee Professor of Government & the Press at Harvard’s Kennedy School, is the founder of The Journalist’s Resource and author of several books, most recently “Is the Republican Party Destroying Itself?” JR plans to post a new installment of his Election Beat 2022 series every week leading up to the midterm elections. Patterson can be contacted at

Further reading:

Brennan Center for Justice, Voting Laws Roundup,” October 2022.

Brian Knight, “Evaluating Competing Explanations for The Midterm Gap,” 2013.

Andrew Rudalevige, “Revisiting Midterm Loss: Referendum Theory and State Data,” Political Science Research 28, 2001.