Five Lessons from the 2025 Election Results

1. “Unless the president is unusually popular, the party holding the White House will suffer significant losses in off-year elections” is a very well-established pattern, so nothing about the Democratic electoral victories in New Jersey, Virginia, and other states yesterday should be treated as shocking or even unusual. But it’s always valuable to confirm that some of the old rules of politics still apply, even as other aspects of our political world have changed. And for all its normal predictability, this thermostatic backlash effect still seems to catch media commentators and other political professionals by surprise. In part, this is because so much attention during the first year of a presidential term is focused on the failure and subsequent internal dissension of the opposition party, making what should in fact be a fully expected rebound of its electoral fortunes seem like a dramatic plot twist.

2. One of the reasons why Trump has been treated as politically stronger, and the Democrats politically weaker, in his second term than in his first has been that we haven’t seen as visible a popular oppositional movement emerge in 2025 as we did in 2017. Yes, there have been anti-Trump protests, marches, and hashtags, but nothing to the scale of what happened eight years ago, and major institutions like universities and corporations have taken a more accommodationist tack to conservative populism than they did in the peak “Resistance” era. But yesterday’s results demonstrate that 2025’s less prominent expressive opposition on the left does not necessarily indicate, or produce, less successful electoral performance for Democratic candidates—or even lower levels of voter participation. This raises the fascinating, and I think quite unresolved, question of how much self-styled “movements” actually affect the outcomes of general elections for top political offices.

3. Based on exit polls and results in particular localities, it’s clear that both Mikie Sherrill in New Jersey and Abigail Spanberger in Virginia performed much better than Kamala Harris in 2024 among Hispanic and Asian-American voters, either matching or exceeding Joe Biden’s performance in 2020. There are three plausible interpretations of this change: (a) minority voters’ growing support for Trump does not—at least yet—apply to other Republican candidates when he isn’t on the ballot; (b) many have become “nature-of-the-times” swing voters who punished Democrats for disappointing stewardship of the national economy in 2024 and are now punishing Republicans for the same thing; (c) Republican gains among socially conservative racial minorities between 2016 and 2024 are now being reversed, perhaps because of the more aggressive immigration enforcement policies and rhetoric of the second Trump term. We’ll need more data and more elections to better understand the causal factors behind this trend (assuming it continues).

4. Zohran Mamdani is poised to become a well-known national figure, in part because both supporters and critics will have a strong incentive to provide him with generous publicity. For various reasons, however, the mayoralty of major cities is normally a political ticket to nowhere—even for politicians who aren’t constitutionally ineligible to run for president. Mamdani is the latest in a number of economically and socially progressive candidates to win election in big northern cities, which is an important development in its own right but a long way from indicating what the “future of the Democratic Party” might be. Mamdani has undeniable political talent, but was also blessed with competing against several unusually unappealing opponents—beginning with an incumbent mayor whose four years of flagrant blundering opened the door to his candidacy. The governing choices he makes will be fascinating to watch.

5. Perceptions matter in politics whether or not they match reality, and the perception that yesterday’s results constituted a “blue wave” (in part because expectations were set by polling that underestimated Democrats’ performance) is likely to have important consequences for a range of topics from the resolution of the government shutdown to candidate recruitment in the 2026 midterms. But it’s as easy to overstate the magnitude of Democratic victories this year as the Republican victories of last year, and indulgence in the former is not the ideal response to indulgence in the latter. The best evidence indicates that we remain in an era of polarized and closely matched parties, meaning that the emotional response of attentive partisans on both sides rises and falls much more dramatically than the actual strength of the parties themselves. And no election is ever the final word on anything; in this country, there’s always another one right around the corner.

2017 Election Recap: Some Rules Still Apply

It's an unfortunate quirk of the American political calendar that Virginia is one of only five states that do not select their governor at the same time as either presidential or congressional midterm elections. Political junkies are especially starved for election news in odd-numbered years, raising the danger of over-interpreting the scattered results that do occur. And the fact that one of the two states happening to vote in 2017 is also home to most of the national political media guarantees that attention will disproportionately fall on Virginia, dependably exaggerating the importance of the outcome of its state-level elections. (It's hard to imagine, say, Missouri or Arizona receiving nearly this much notice if it were in the same position.)

The real importance of the Democratic victories in Virginia (and New Jersey, and in scattered locations elsewhere) is not that they tell us much that we don't already know. Instead, they are just a few more data points confirming one of the most dependable rules in American politics: the president's party predictably loses ground in down-ballot races, and the amount of ground lost is correlated with the president's unpopularity.

And Trump is quite unpopular. His job approval rating is the lowest of any president on record during his first year in office, and roughly half of the country not only disapproves but strongly disapproves of his performance. The unexpectedly high Democratic turnout in Virginia tonight suggests that Democratic voters are highly motivated to express their opposition to the president and the ruling GOP when given the opportunity—even in the absence of a personally charismatic Democratic candidate. This is hardly a surprise, but it still demonstrates that the laws of American politics have not been completely rewritten in the Trump era. There were many good reasons to believe that 2018 was shaping up to be a good Democratic year before tonight's results rolled in, and the outcomes in Virginia and elsewhere simply constitute one more good reason.

If there is any actual further importance to the outcome tonight, it comes from the effect that the election returns might have on the calculations of incumbent politicians and potential challengers. If a number of Republicans representing competitive districts begin to mull retirement in response to tomorrow morning's headlines, and if an array of talented Democratic candidates gain enough confidence from these results to throw their hats into the ring for Congress and other offices, than tonight's events will indeed exert an effect on what happens next year. But there is still a long way to go until November 2018, and the nation's political climate cannot be foreseen with precision a full year ahead of time on the basis of a few state and local elections tonight.