Tuesday night could hardly have gone better for Democrats. They easily held the governor’s mansion in New Jersey, flipped it in Virginia, and won voter approval to gerrymander California aggressively in their party’s favor. In all three of these elections, they won by more than Kamala Harris did in the 2024 presidential election, and by a much larger margin than the polls had suggested they would. Perhaps Democrats’ most impressive victory was in Georgia, where they ousted two Republican members of the statewide energy-regulation board by 25 points.

These are serious achievements for a party in the wilderness. But to conclude that Democrats have solved their electoral difficulties, or have even begun to, would be a mistake. They are still in deep trouble.

In the Trump era, Democrats have routinely dominated low-turnout elections. The more random the election date and the more under the radar the office, the better they seem to perform. Getting the highly engaged to vote blue has never been the problem. Democrats’ weakness manifests in presidential elections, in which more people vote, and in elections for the Senate, where small, rural states hold disproportionate power. Persuading an ever more engaged slice of the electorate to vote for you by an ever higher margin is better than nothing, but it does very little to help rebuild a party at the national level.

Democrats’ edge in low-turnout elections might even be growing. In special elections in 2017 and 2018, Democrats outperformed Hillary Clinton’s 2016 numbers by an average of six points. In the first 30 special elections since President Donald Trump’s inauguration, they outperformed Harris by an average of 13 points.

Maybe Democrats are overperforming because Trump has gone out of his way in his second term to infuriate and activate them. But these shifts also reflect changing coalitions. Since Trump was first elected, the Democratic base has gotten older, whiter, richer, more female, and more highly educated. For decades, those have been the voters that most consistently head to the polls. Early data do suggest that a small but meaningful portion of Democratic voters on Tuesday were Trump voters in 2024. That’s an encouraging sign for Democrats if true. But it is also consistent with the possibility that the party has increased its lead only with highly engaged voters, without having fixed its irregular-voter problem.

[Marc Novicoff: Democrats don’t seem willing to follow their own advice]

All things being equal, you’d rather have a high-propensity coalition. You don’t have to worry so much about getting out the vote, and building power at the local level is easier, whether on the school board or in the state legislature. But having the high-propensity coalition comes with risks. The most obvious is that it’s less helpful in winning the presidency. Presidential elections attract many more voters, including the kind who don’t vote in midterms or off-year governor elections. Those irregular voters broke for Trump in 2024. Neither party seems to have internalized this fact. Democrats keep working hard to make voting easier, and Republicans push paperwork requirements that at this point might keep more of their own voters from casting a ballot. (Whether irregular voters’ rightward shift is a Trump-specific phenomenon is unclear, and will remain so until another candidate heads the Republican presidential ticket.)

The other risk of a high-propensity coalition is that you misread your voters’ enthusiasm as evidence that you have a lot of them in a lot of places. In 2017, Democrats were overjoyed as their nominees for governor trounced their Republican opponents in Virginia and New Jersey. The next year, Democrats kept their momentum and took back the House in a so-called blue wave. But in the Senate, they lost seats: Democratic incumbents were defeated in North Dakota, Indiana, Missouri, and Florida. The Democratic base had certainly become more enthusiastic, but it had also become more concentrated. The dynamic has only accelerated since then. Democrats lost Senate seats in Montana, West Virginia, and Ohio last year, all three of which they had been able to keep in 2018.

This is the progression that Democrats must halt. If they’d like to ever stop the confirmation of a Trump appointee, they need to have the Senate. If they’d like to ever pass a law without Republican assistance, they need to have the Senate. Yet they have never in modern history looked further from winning it. In 2026, if Democrats manage to hold Senate seats in Georgia and Michigan, where Trump won last year, they will almost certainly still have to beat Susan Collins in Maine, win a Senate seat in North Carolina (a state that Trump has won three times in a row), and pick up two seats in states that Trump carried last year by more than 10 points. The map in 2028 is only marginally more hospitable.

Democrats might lament the unfairness of the fact that small states, which today are disproportionately white, rural, and conservative, hold disproportionate power in the Senate (and a little more power in the Electoral College). But the rules of the game have been in place for 250 years, and only in the past 10 has the Democratic base become so coastal, urban, and liberal that the party seems to have no chance in most of the country. That’s their problem, and no off-year electoral sweep should make them think they’ve solved it.

From The Atlantic via this RSS feed