Early Polls and Off-Year Elections
Republicans are still a mess
Trump and Republicans face strong headwinds going into next year’s elections. In spite of Biden’s poor polling numbers, I remain convinced that Republican extremism and dysfunction make Biden the favorite in 2024.
Joe Biden is not a popular president. A recent New York Times/Siena poll found that Trump is ahead of Biden among registered voters in five out of six swing states. Biden’s numbers have been historically bad throughout his term. His net approval is currently at around -17 points, which is further under water than Donald Trump was at this point in his presidency. It’s clear from the polling that voters blame Biden for the challenging economic conditions of the last few years, and have doubts about his ability to be an effective president given his age.
Just two days after that New York Times/Siena poll came out, Democrats defeated Republicans almost across the board in last Tuesday’s off-year elections. Democrats retained control of the state Senate and won back control of the state House in Virginia. They held onto their substantial majority in the state Senate and increased their majority in the state Assembly. They expanded their control of the state legislature in New Jersey. In Pennsylvania, Dan McCaffery won an open seat on the state Supreme Court, giving Democrats a 5-2 majority. In Kentucky—a state Trump won by 26 points in 2020—Democrat Andy Beshear won reelection as governor. In Ohio—another state Trump won—voters approved a measure adding the right to obtain an abortion to the state constitution by a substantial margin. Democrats’ one high profile loss came in Mississippi, where Brandon Presley failed to win the race for governor, but nevertheless managed to do well for a Democrat in Mississippi. These off-year results are not outliers. Democrats have outperformed expectations in special elections all year and did well in almost every major election since Trump’s victory in 2016.
What explains this apparent disconnect? Nate Cohn argues there’s no real contradiction: voters broadly support Democrats and liberal causes, but disapprove of Biden. Some polls also suggest that lower-propensity voters may now favor Republicans, so it’s possible Republicans will do better in 2024 when these less engaged voters turn out in larger numbers. I think Biden’s poll numbers probably do reflect how voters feel; they really aren’t enthusiastic about his candidacy. But normally you’d expect the party out of power when there’s an unpopular president to have the turnout advantage. I’m skeptical that voters’ feelings about Biden are that disconnected from their feelings about the Democratic Party in general; voters who prefer Democrats to Republicans will mostly still vote for Biden in 2024.
Political scientists have found that presidential approval ratings and economic conditions are somewhat predictive of presidential election results, although the relationship is weak this far ahead of the actual election. Ronald Reagan, Bill Clinton, and Barack Obama all had negative approval ratings at this point in their presidencies and went on to win reelection. None of them had approval ratings as low as Biden’s, but Biden’s likely opponent is the one president who had a similarly bad approval rating. Last week’s elections show that how voters feel about Biden and about the economy isn’t stopping them from voting for Democrats.
Nor are head-to-head polls like the recent New York Times/Siena poll that show Trump beating Biden likely to be very meaningful. Head-to-head polling seems to have essentially no predictive value this far in advance of an election. That doesn’t mean the polls are necessarily inaccurate. But while they may provide an accurate snapshot of how voters currently feel, who voters say they’ll vote for now is not a good indicator of who they’ll actually vote for a year from now. Early polls tend to be a referendum on incumbents, rather than a serious choice between candidates. They allow voters to express their dissatisfaction in a way that has no real consequences. Trump isn’t the source of voters’ frustration with the current government; as long as the prospect of him returning to power seems far off, he’s mostly a sideshow.
But Trump’s baggage will become more salient as the general election gets underway in the spring. It’s a lot of baggage. He currently faces 91 felony charges in four separate trials and will be running for office while he is literally out on bail. He’s credibly accused of attempting to stay in power illegimately—the most serious civic crime a democratic official can commit—and is likely to be convicted of at least some of the charges against him. That somehow won’t hurt his popularity with his base, but as the election approaches it’s likely to drive away voters he’d need to win the election. Trump’s closest allies have likewise underperformed badly in competitive districts over the last few elections. Republicans meanwhile have been unable to agree on a platform that satisfies primary voters and has any hope of appealing to voters in swing districts. Their dysfunction has been on display in the House, where they’ve barely even been able to agree on a Speaker. Instead, they’ve been trying to ram through wildly unpopular policies like abortion bans—which have been defeated virtually everywhere they’ve been on the ballot—while pushing bizarre conspiracy theories and manufactured culture war issues.
Biden certainly has problems of his own. He will have to allay concerns about the economy and about his age. He will need to to address the renewed conflict and ongoing humanitarian disaster in the Middle East without somehow alienating some of his supporters. But Biden remains the incumbent and the leader of a popular party. The large electoral college advantage Republicans have enjoyed in recent elections seems to have lessened. Inflation has fallen while growth remains strong. Trump’s own age and his apparent lapses will make it hard for him to capitalize on concerns about Biden’s age. Abortion access will be at stake again in key states in 2024. Right now, Good Judgment gives Democrats a 65% chance of winning the presidency in 2024, while Metaculus gives Biden a 59% chance of winning a head-to-head race with Trump. If the 2024 election were simply a referendum on Biden, he would definitely be in trouble. But with voters in the US primarily motivated by their opposition to the other party, the strength of Biden’s support may matter less than the strength of the opposition to Trump. My bet is still that when the Trump show returns to center stage, voters will turn out again to vote against him. My forecast remains unchanged since the last time I wrote about the 2024 presidential election.
My Forecast
70% chance the Democratic nominee will win the presidency in 2024
You can see all my recent predictions on Fatebook here.
Thanks again to everyone who has already supported my work by buying a paid subscription! My ability to continue doing this depends on you. If you’re interested international politics, you might enjoy the recent substack posts from my forecasting colleague on Taiwanese readiness. And if you enjoyed this post, please share it with others!
A note of caution on the Virginia election: Although they remained in control, Democrats actually lost a seat in the Senate, from 22 to 21. In the house a few seats flipped, but that could be attributed more to non-partisan districting replacing the previous districts that were favorable to Republicans. A much higher number of seats have been flipped in past elections. I was expecting more of a blue wave, and actually look at the results as a signal against one in 2024. It should be enough for Democrats to get a few key ballot measures (abortion, voting rights) on the ballot, though, which theoretically helps Democratic turnout in 2024.