Down to the Wire
Control of Congress may come down to a few seats
Democrats’ prospects have improved dramatically since Kamala Harris replaced Joe Biden as the presidential nominee. But the race for Congress is likely to be very close, and an extremely challenging Senate map means Democrats are likely to lose the Senate.
Election Day is less than three months away. Early voting begins in some states at the end of September, so there’s not a lot of time left for candidates to make their cases. The Democratic National Convention this week will kick off what are likely to be the most crucial weeks of the election. With Joe Biden stepping aside—it’s now clear that Biden was a heavy drag on the Democratic prospects—Kamala Harris is the favorite to win the presidency. But the election will be close and control of Congress could be decided by just a few seats.
FiveThirtyEight currently gives Democrats a modest lead of 1.1 points on a generic congressional ballot. Generic ballot polls ask potential voters which party they prefer rather than about specific candidates, so they’re a rough indicator of the national party preference. But Democrats’ lead on the generic ballot doesn’t necessarily mean they’re favorites to control Congress. Since 2002, Republicans have won control of the House when Democrats’ lead on FiveThirtyEight’s final generic ballot average has been less than about 2.5 points. That’s because FiveThirtyEight’s generic ballot average has generally—but not always—overestimated Democrats’ final share of the popular vote by several points, and because Democrats probably need to win a little more than half the popular vote to win control of the House.
Over the next three months, the generic ballot could shift a couple of points in either direction—I weakly think it will continue to trend toward Democrats—but all being equal history suggests Democrats need around a 5-point lead on the generic congressional ballot to be favorites to win the House. Congressional handicappers nevertheless think the race will close. According to 270toWin’s consensus forecasts, Republicans are the favorites in 209 districts, Democrats are the favorites in 205 districts, and 21 districts are tossups. One party could plausibly win most of the tossup districts—and win control of the chamber by a comfortable margin—but right now it looks like House control will be decided by just a few seats.
Control of the Senate, meanwhile, depends less on the general national vote than it does on the particular seats and the particular candidates up for election. As I wrote a year ago, Democrats will have a hard time keeping control of the Senate in 2024. Democrats are defending 23 of the 34 Senate seats at stake this year, which means they have twice as many seats to lose. They’re defending 7 seats in states that Donald Trump either will definitely or could plausibly win, while Republicans don’t have any seats in states Harris is likely to win to defend. With Joe Manchin stepping down, Democrats are certain to lose West Virginia. Democrats will lose control of the Senate if they lose even one more seat—unless they can somehow pick up Florida or Texas—so they basically have to run the table everywhere else. Democrats are solid favorites to hold on to most of their contested seats, in part because Republicans are running a number of weak candidates again this year. Control of the Senate is therefore likely to come down to whether both Sherrod Brown and Jon Tester can hold on to their seats. Brown has a good chance to hold on in Ohio, but Tester is trailing in Montana.
With Harris’ surge in the polls, Split Ticket’s model gives Democrats a 58% chance of retaking the House—projecting them to win a razor-thin 3-seat margin—but just a 16% chance of keeping control of the Senate. Metaculus forecasters give Democrats a 56% chance of retaking the House but just a 28% chance of keeping control of the Senate. Polymarket bettors give Democrats a 62% chance of retaking the House but just a 30% chance of keeping control of the Senate. I’m more skeptical that Democrats can retake the House given their small lead on the generic ballot, but I think they probably have a small edge that hasn’t yet shown up in the polls. Democratic control of the Senate probably hinges on whether Tester can somehow win in Montana—because if he can win there Brown can probably win in Ohio—even though Trump is likely to win Montana by double digits. That race is close, but Tester is a clear underdog.
My Forecasts
54% chance Democrats will win control of the House in 2024
25% chance Democrats will win control of the Senate in 2024
You can see all my predictions or add your own on Fatebook here.
If you’re a US citizen, you can register to vote or make sure you’re registered to vote at vote.gov/register/. I plan to record new podcasts soon, but if you’d like to hear me talk about forecasting and existential risk on other podcasts, I put together a playlist of interviews I’ve done on Spotify here. Telling the Future depends entirely on reader support, so if you enjoyed this newsletter please consider supporting me by buying a paid subscription.
