To what extent will Biden’s low approval explain the midterm results?


This tweet from FiveThirtyEight’s Nathaniel Rakesh was mostly intended as a joke.

Note the date. He posted it on November 6, 2020, the day Donald Trump’s re-election loss became apparent but even before it was formalized. At the time, more than two years ago, Rakesh was doing some kind of tongue-in-cheek anticipation: Because first-term presidents always saw their party losing seats in their first midterm elections, a Joe Biden win in 2020, heralded a Democrat loss in 2022. .

This may have certainly been a mistake. Biden would have been extraordinarily popular and the recovery from the pandemic was extraordinarily strong. A lot of things can happen over the course of a couple of years that fluctuate expectations. But that very expectation, that Republicans will do better in the midterms than Democrats, seems almost inevitable with the final votes counted.

Asking a question that candidates, consultants, and many observers prefer not to ask: How far has the midterm election result been for most of the past two years?

Subscribe to How to Read This Chart, Philip Bump’s weekly data newsletter

It’s the afternoon of November 8, 2022, as I’m writing, which means we haven’t done that yet I know What will the midterm elections bring? But we can look at what happened in the past midterms to understand.

FiveThirtyEight has historical approval rating averages that allow us to compare apples to apples here – or at least, as close to apples as might be expected. If we look at the past 100 days leading up to the first midterm elections for a new president in office (excluding those who took office under unusual circumstances), we see some patterns emerging.

The first is that there is much less variance on average now than there was previously. This is largely a function of polarization: parties love the bosses of their own party and hate the bosses of the opposing party. We also see that the president whose pre-midterm approval pattern matches Biden best is Ronald Reagan, something the president might find encouraging when he considers the 2024 bid.

But what we often see is that the midterm level of support for the president’s party (indicated by a horizontal dotted line) does not correlate well with approval. After all, the lead-up to 1994 was the period when Democrats consistently held a large majority in the House of Representatives even if they weren’t in the Senate or if Republicans controlled the White House. This is the main reason that the Democratic National Assembly vote in that top row (i.e. in 1962 and 1978) is above 50 percent, while the Republican vote (in 1954 and 1970) is below 50 percent—even though the Dwight D. Eisenhower was more popular on Election Day in 1954 than Jimmy Carter was in 1976.

So let’s look instead at the change in seats held by the president’s party at each half-term. Here we see a more obvious connection.

In only one election – 2002 – did the president’s party win midterm seats. This was largely a result of George W. Bush’s popularity, which was boosted in the aftermath of the September 11, 2001 terrorist attacks. In each of the other nine elections since 1954, the president’s party has lost seats. And the less popular the president, the greater the loss.

The trend line shown there shows how approval correlates with seat loss or gain, but it is not predictive. For example, not so because because Biden’s average approval falls north of 40 percent (shown by the horizontal blue dotted line), we can expect Democrats to lose more than 60 House seats. Other factors come into play, including how many seats have been determined to be safe for one party or the other.

In the two previous elections in which presidents were as unpopular as Biden, 1982 and 2018, their party lost dozens of seats. The Democrats’ loss of up to a dozen this year means the Republicans’ control of the House of Representatives. In other words, looking Just With Biden’s approval, we expect the GOP win that Rakesh jokingly predicted. Even if Biden’s approval is higher, in fact, his party’s narrow majority in the House means Republicans are likely to win, given the precedent set out above.

There will be a lot of discussion over the next few weeks about what happened and why. Candidates will insist that it was their specific qualities that led to their victories; Advisers will insist that these were their clever tactics. This is what happens.

The truth is that certain patterns occur in politics over and over again. Someone might point out that Democrats are having a bad Tuesday night — as predicted two years ago.

Leave a Comment