Ticket splitting is dead. National parties are now everything.

Every single state that elected a Republican senator this November voted for Donald Trump — and every single state that elected a Democratic senator voted for Hillary Clinton.

That’s a first in American history — at least going back to 1913, when the Constitution began mandating the direct popular election of senators. And it’s a dramatic reversal from much of the middle of the 20th century, when voters frequently backed senators of one party while also supporting the opposing party’s presidential nominee — a phenomenon known as “ticket splitting.”

The chart above, produced by the University of Virginia’s Center for Politics, shows what percentage of states voted for the same party’s presidential and Senate nominees. (The green line represents the percentage of states backing both a senator and presidential candidate of the same party; the orange line shows the percentage of states that split between parties’ Senate and presidential candidates.)

The finding confirms a long-running trend: that ticket splitting is now virtually dead. The key culprit appears to be political polarization: Whereas the parties in the 1960s and 1970s were similar enough that a candidate’s local popularity could make a great deal of difference to a state’s voters, the Republican and Democratic caucuses are now so divided on the issues that it makes little sense to, say, back Hillary Clinton but oppose Democratic Missouri Senate candidate Jason Kander.

In an interview, the Center for Politics’ Geoffrey Skelley adds that the vote share of the parties’ presidential candidates and Senate candidate were also extraordinarily closely tied together this year (a 90 percent correlation). In other words, basically no Senate candidates did much better or much worse than their party’s nominee.

“There aren’t many split-ticket voters in this day and age,” Skelley says.

We have an idea of American democracy that goes something like this: The Constitution gives different politicians oversight over different governmental bodies, which in turn affect citizens at the national, state, and local levels.

The politicians who control these different governmental bodies are, naturally, different people. So if a state lawmaker is doing a crummy job or passing unpopular legislation, her constituents can punish her individually at the ballot box.

The problem is that this is not what happens at all. Mounting evidence suggests that the entire party’s fate hinges on the popularity of its president or presidential candidate, and that the rest of its fortunes will rise or fall on pretty much that person alone.

That may sound like an exaggeration. But take this study by professor Steven Rogers of St. Louis University from earlier this year. It found that the single most important factor in state legislature races is what the voters think of the sitting president — even if the president has essentially no hand in setting statehouse policy.

Voters are about 6 percent more likely to vote against their state lawmaker if they disapprove of their state legislature, about 9 percent more likely to do so if they disapprove of their governor, and a full 40 percent more likely to vote against the president, according to Rogers.

There’s ample evidence this dynamic has a powerful effect at the congressional level as well. Once upon a time, it was possible for parties to get their clocks cleaned at the presidential level while also adding to their control of Congress and the statehouses.

Back in the 1960s and ’70s, for instance, most elections saw about 30 percent of people vote for one party at the presidential level and a different party down ballot, according to political scientist David Kimball of the University of Missouri St. Louis. In 1984, President Ronald Reagan won a historic landslide while Democrats racked up big gains in the House while also running evenly with the GOP in the Senate.

That’s all changed. In 2012, fewer than 10 percent of their voters split their tickets — the lowest number ever, according to Kimball. We don’t have the exact statistic yet on this front for 2016, but Skelley says it’s likely to prove much the same story.

The fact that all politics is national may have helped made it difficult for congressional Democrats in battleground states to do much better than their party’s presidential nominee. But it also suggests that — if Donald Trump’s record unpopularity holds — they’ll have a real opportunity in the midterm 2018 elections.