The 2023 elections were a victory for liberal governance. Democratic Governor Andy Beshear was re-elected in Kentucky. Ohio voted for enshrine the right to abortion in the state constitution. Democrats took control of both houses of the Virginia legislature And strengthened their hold on New Jersey. The main positive point for the Republicans was that Republican Governor Tate Reeves won re-election in Mississippi.
Yet it seemed like the only thing people cared about was what the 2024 election results portend. I have the impulse – I was one of those people who I read too much in these off-year election results. But after a Republican ride to the 2021 elections having failed to predict a mixed but generally good 2022 midterms for Democrats, I took a closer look – and found that off-year election results are not the best predictor of future elections . As a result, it’s unclear what, if anything, we learned last week about 2024.
Kentucky and Mississippi are the worst oracles
Let’s start with the easiest narrative to debunk: the idea that the results of the Kentucky and Mississippi gubernatorial elections tell us something about the 2024 presidential election. After Beshear’s victory, some observers have noted the fact that, since 2003, the party that wins the Kentucky gubernatorial race has always won the White House the following year. But you should be wary of statistics like this that claim something “always” happens or that rely on binary outcomes rather than margins. For example, the 2019 Kentucky gubernatorial race was extremely close and could easily have gone either way. If just 5,137 more people had voted Republican in this election, Kentucky’s predictive streak would have been broken.
Instead, let’s take a look at margins in Kentucky gubernatorial races (and, for good measure, Mississippi gubernatorial races) since 1999 and how they compare to the national popular vote margins for the following year’s presidential election . It turns out there is virtually no correlation.
This should come as no shock to anyone who followed this year’s elections: Democrats won in Kentucky and came close in Mississippi, because they had strong candidates with multi-party appeal. President Joe Biden, meanwhile, is unpopularaccording to 538’s tracking of his approval rating, and likely won’t be competitive in deep red states like these in next year’s presidential race.
Take New Jersey and Virginia with a grain of salt
Next: Are the legislative election results in New Jersey and Virginia predictive of future elections? I used to believe they were good doppelgangers for federal House elections, at least: a large sample of elections in a uniform campaign environment on a regular date with a good mix of incumbents standing. running for re-election and vacant seats. But when looking at their historical records, New Jersey and Virginia appear to be unreliable bellwethers at best.
After the 2017, 2019 And 2021 elections, I compared the Democratic or Republican margin of victory in each legislative district in New Jersey and Virginia that was actively contested by both parties to the baseline partisanship of that district – essentially, how we would expect this district votes whether the national political environment was tied among Democrats. and Republicans.* I then averaged these numbers to calculate the average outperformance of Democrats or Republicans in each chamber each year. If this chamber’s elections were effective in predicting future elections to the U.S. House of Representatives, this outperformance figure should be closer to the national popular vote for the House the following year. But since at least 2017, that hasn’t really been the case.
Certainly, the results from New Jersey and Virginia seem to be moving in the right general direction. The year with the biggest Republican outperformance was 2021, and 2022 was the only time in the last three cycles that Republicans won the popular vote in the House. Likewise, the year with the greatest Democratic overperformance (2017) preceded the cycle with the greatest Democratic popular victory (2018). But exact overperformance doesn’t come close to getting close to the eventual popular vote in the House — just look at 2017-2018, when the Virginia House missed by 7 percentage points. Additionally, this is a fairly small sample of elections, so we cannot be sure that the directional trend we think we observe will persist over a longer period of time.
According to this method, on average, the 2023 New Jersey State Senate election results The results were essentially neutral, while Republicans outperformed by about 3 points in the Virginia State Senate and House of Representatives. In other words, despite the fact that Democrats won both houses in Virginia, Republicans actually had a strong election night once you account for Virginia’s blue hue (although, as pointed out Kyle Kondik of Sabato’s Crystal Ball, Virginia tends to vote more against Republicans than for the president). Democrats won every chamber by the smallest possible majority – one seat (21-19 in the Senate, 51-49** in the House). If the directional trend we identified above is true, perhaps that means 2024 will fall somewhere between 2020’s 3-point Democratic victory in the popular vote in the House and 2022’s 3-point Republican victory But again, these numbers have been such imprecise measurements in the past that all it really tells us is what we already knew: Either party could win in 2024.
Trust in special elections?
Ultimately, the most predictive 2023 elections may be among the most understated. On November 7, special elections were held in Rhode Island’s 1st Congressional District as well as seven state legislative districts (although one of them was undisputed). And as we have written several timesA party outperformance in special elections (using the same methodology I used for New Jersey and Virginia) a historically, it was rather predictive of the popular vote of the House in the following general elections.
And the results of those special elections were mixed, although overall they were better for Democrats than Republicans. Democrats outperformed baseline partisanship in four of seven elections, and the average outperformance on November 7 was D+3.
This was actually a departure from what we were told in the pre-November special election. Between January and September, Democrats performed spectacularly in the special elections, yielding margins that were, on average, 11 points bluer than baseline partisanship. We don’t know if the Democrats’ more moderate performance on November 7 was due to a change in the national environment or if it was just noise in the data, but it is at least the case. interesting that it was more in line with generic congressional ballot polls (which show absolutely no sign of a D+11 national environment) and with the results from New Jersey and Virginia.
That said, it is special elections throughout the cycle that have historically proven to be predictive, not just special elections on odd election days, and with the November 7 special election taken into account, Democrats still outperform base partisanship by 9 points on average over the year.
So, after all this analysis, I fear we are not much closer to answering the question of what last week’s election will mean for future elections. Some 2023 election results clearly tell us nothing about 2024; others may tell us something, but we can’t really be sure; still others have been historically very useful but are pointing in confusing directions this year. In other words, I don’t recommend changing your outlook for 2024 based on what happened last week.
Instead, consider the 2023 election for what it was: a series of Democratic and liberal victories that were important in their own right. More than 25 million people live in Kentucky, Mississippi, New Jersey and Virginia and will be directly affected by the results. Ohio has guaranteed the right to abortion for 2.6 million women of childbearing age and legalized marijuana for 8.7 million adults. In Pennsylvania, the Democrats won key victories in judicial and county-level races this will be important for the 2024 presidential race, not because of what they portend, but because they will affect how this election will be organized logistically and how ballots will be counted. These things are the real legacy of the 2023 elections.
*A district’s “base partisanship” is the average difference between how the district voted and how the country voted as a whole in the last two presidential elections, with the most recent presidential election having a weighting of 75 percent and the second most recent presidential election, a weighting of 25 percent. This is admittedly an imperfect exercise in Virginia, where mail-in ballots were not allocated by precinct until very recently, which meant that presidential results by legislative district were only an estimate. However, Daniel Donner of Daily Kos Elections made a similar analysis which compared the statewide popular vote for the Virginia House to the commonwealth’s baseline partisanship, and it came to the same conclusion.
One final note: I didn’t bother to do this exercise for the elections to the New Jersey Assembly, the state’s lower house, which also take place in odd years. Assembly districts each elect two legislators, and all candidates in the district run in the same race (the top two receiving votes are elected). This irregular arrangement makes Assembly elections too different from typical elections to be analyzed reliably.
**One Virginia House district, the 82nd, still does not have a projected winner according to the Associated Press. However, Republican Kim Taylor 78 votes ahead is probably big enough to withstand a recount.