President Trump turns to House Speaker Nancy Pelosi (D-Calif.) at the Capitol on Feb. 5 during his State of the Union address. At left is Vice President Pence. (Doug Mills/AFP/Getty Images)
Opinion writer

If you were a moderate Democrat elected to Congress in 2018, you might worry about how the sweeping proposals from your more liberal colleagues will reflect on you. Are the voters in my swing district going to think I’m some kind of far leftist, too? Should I make efforts to signal my centrism so they don’t turn against me? Will that work?

The answer, according to some fascinating new data, appears to be no. In fact, there is a strong case to be made that in congressional elections — especially for the House — the general election barely matters anymore. All the action is in the primary. Once that’s over, you can be a liberal Democrat or a moderate Democrat, and you’ll do just as well.

That goes against what political professionals have believed pretty much forever. Members of Congress spend a good deal of time worrying about how this or that vote will be received back in their districts, and about whether they’ve constructed an ideological profile that matches their constituents.

We see that struggle in articles such as this one about moderate freshman Democrats being confronted with statements made by their colleagues and proposals such as the Green New Deal, as they worry about whether they’ll be dragged too far left.

That seems perfectly rational, and for a long time it was. If you were a moderate in a general election, you could hold on to your own party’s voters and poach some voters from the other party, too. If you utilized the power of incumbency, you could forge a bond with voters that would be stronger than party, at least in enough cases so you could be reelected — even if your party’s presidential candidate didn’t win your district.

But that’s no longer true. Emory University political scientist Alan Abramowitz shared with me an analysis he did of the effects the district’s presidential vote had on House elections, whether there was an incumbent running and whether that incumbent is more or less ideologically extreme.

Here’s what his results suggested. Through the 1990s, while there was still a fairly strong relationship between presidential and House votes, the other two factors mattered a lot, too. Incumbents did far better than non-incumbents, even controlling for their district’s partisanship. And being more moderate was a big help.

But by the time you hit the 2008 election, ideology had stopped making a difference. According to the analysis, a moderate or a far-left liberal will do equally well if their district looks roughly the same. And by 2016, the effect of incumbency was reduced from a huge benefit to a tiny one.

“Incumbents still get reelected at a very high rate, but the reason is that the large majority of them are in districts that favor their party,” Abramowitz told me.

In 2018, Abramowitz looked at whether Democratic candidates were endorsed by Our Revolution, an organization affiliated with Sen. Bernie Sanders (I-Vt.), as a proxy for highly liberal ideology. He found that they did no better or worse than more moderate Democrats, once you took the tilt of their district into account. Their ideology made no difference. Ticket-splitting is largely a thing of the past.

So what does that mean if you’re a member of Congress? Let’s say you’re a Republican representing a district that tilts slightly Democratic, such as former representative Barbara Comstock (R-Va.). You can make all kinds of efforts to signal to voters that you’re a moderate but, come Election Day, you’re going to lose just as surely as you would if you had been a hard-right conservative, which is just what happened to her in 2018 when high Democratic turnout doomed her.

“It doesn’t do you any good,” Abramowitz says, “to position yourself in the center in hopes that that’s going to attract more votes from the other party. It just doesn’t seem to work.”

In a way, voters have stopped caring so much about their members of Congress as individuals. “What’s happening is that people are not voting based on who they want to represent their district,” says Abramowitz. “They’re voting based on which party they want to control the chamber.”

Which is a perfectly rational thing for a voter to decide. If control of the chamber really is at stake, even if you think the candidate from the other party is actually a pretty good guy, it would be foolish to support him or her if it means your party won’t be in control.

What’s really striking is how closely votes for the House and votes for president have locked into alignment. Abramowitz notes that during the 1960s, the 1970s, and even the 1980s, the correlation between the presidential vote in a district and the House vote in that district was about 0.6 — substantial, but not overwhelming (correlations run from 0, meaning no correlation, to 1, meaning perfect correlation). In 2016, it was .97, higher than it had ever been before. The candidates who thought they could overcome their district’s fundamental partisanship by constructing a more moderate profile were almost all on a fool’s errand.

To be clear, this doesn’t mean there won’t be a few outliers who manage to pull it off. But in most cases, the general election will turn out the same way no matter who the nominees are.

So how did this happen? It has to do with the increasing ideological coherence of the parties, which provide evermore distinct agendas. The rise in recent years in negative partisanship — feeling stronger about your hatred for the other party than you do about your affection for your own party — also plays a key role in reducing split-ticket voting.

But there’s a vital media story here, too. As local newspapers have died off or been taken over by large corporations, coverage of local politics across the country has been hollowed out. At the same time, cable news and the Internet have given people access to endless discussions about national politics, national figures and national controversies. These encourage voters to think about their local candidates not as distinct individuals but as representatives of their parties.

All of which lends further weight to the idea that mobilization is becoming much more important than persuasion. As much as we might prefer a politics in which everyone has an open mind and candidates can win over voters from any party with a compelling enough argument, that’s not how things work today. And it’s hard to see how they could ever go back.

Read more from Paul Waldman:

What will Trump offer the voters next year? Probably not much.

Why Democrats need to be ready to kill the filibuster

In 2020, we may finally get to talk about what kind of capitalism we want

There’s yet another Trump administration scandal brewing. And it’s a doozy.