Welcome to Second Rough Draft, a newsletter about journalism in our time, how it (often its business) is evolving, and the challenges it faces.

Subscribe now

Everyone knows by now that the Democrats did better than expected in the midterm elections. How did the press do?

This was the first election cycle for widespread adoption of a “democracy beat,” coverage expressly devoted to the challenges posed to our system by, among other things, election denialism and the aftermath of the January 6 coup attempt. I don’t think it’s too soon to take stock of what we’ve learned—or might learn—from this approach, and about political coverage generally, particularly as attention to the presidential campaign of 2024 begins.

Here are a few tentative conclusions:

In a closely divided electorate, look more at the deciders and less at the divisions

We’ve known for more than six years now that the country is closely divided on a partisan basis, and an extraordinary amount of attention has been focused on the divide. Last week’s results suggest to me that our political coverage would be much more insightful if it instead focused more on the people who will necessarily decide elections in such a circumstance.

Let’s review examples of what this means, based on the latest results:

In Wisconsin, the Democratic governor, Tony Evers, was re-elected by roughly three times his margin in 2018. But the same voters declined to replace Republican Senator Ron Johnson with Evers’s lieutenant governor, Mandela Barnes. In a state Joe Biden carried by 21,000 votes two years ago, Evers received nearly 50,000 more votes than Barnes, and Johnson won by 26,000. At least 35,000 people cast ballots for both Evers and Johnson.
In Georgia, Republican Governor Brian Kemp handily won re-election while Democratic Senator Raphael Warnock ran 36,000 votes ahead of his GOP challenger in the first round of voting. At least 100,000 Georgians seem to have voted for both. That number is about eight times Joe Biden’s 2020 margin in Georgia.
In Alaska, both Democratic at-large Representative Mary Peltola and Republican Senator Lisa Murkowski seem likely to have been re-elected when ranked-choice tallying is complete. Perhaps as many as 80,000 Alaskans— more than one-third of the electorate— may have ranked both as their first choice.
Ditto New Hampshire, with the re-election of both Republican Governor Chris Sununu and Democratic Senator Maggie Hassan by large margins, where the overlapping vote appears to have accounted for more than 10% of all voters.

getting Sununu and Hassan in the same frame IS possible: AP did it here

There will surely be more examples to come. In Nevada, Jon Ralston, one of the nation’s savviest political reporters, and his colleagues at the excellent Nevada Independent (for which I have done a tiny bit of consulting), have launched a call-out seeking to understand Nevada ticket-splitters, as the state re-elected Democratic Senator Catherine Cortez Masto, but chose Republican sheriff Joe Lombardo over the incumbent Democratic governor.

Instead of directing our attention, and our coverage, to the depth of division—constantly contrasting the partisan cross-tabs in every manner of polling-- wouldn’t it make more sense to try to better grasp who these split-ticket voters are? Instead of endless pieces talking to Trump voters in diners, what we need is to understand the Evers/Johnson, Kemp/Warnock, Peltola/Murkowski, Sununu/Hassan and Cortez Masto/Lombardo voters and why they made the choices they did. These are the people really driving American politics today.

We do this in legislative coverage, with appropriately outsize attention to figures like Sens. Joe Manchin and Kyrsten Sinema, because they are at the locus of the possible in an arena where, as Bismarck said, that is the essence of the art of politics. We should similarly re-orient coverage of voters.

Cut down on the horse-race coverage, if only because it’s embarrassing

From almost the beginning of so-called sophisticated horse-race coverage of political campaigns with Theodore White’s The Making of the President in 1960, there have been pleas to cover pundits and polls less and “issues” more. It’s been a losing battle. People like horse races, especially the sorts of people who become political reporters (as well, importantly, as those who become political operatives, their key sources). But for four consecutive election cycles now, the horse-race coverage has not only been shallow, it’s been wrong, which is like eating off-brand fast food: it’s bad for you, and ends up tasting awful.

Instead of trying to “fix” this, how about if we took steps to protect ourselves from the next embarrassment— to, in effect, pack our lunch at home?

One important step in this direction, I continue to think, would be for news organizations to stop all polling of their own. If they did that, we would at least be spared all of the single-poll stories these surveys generate as every newsroom headlines its own poll. That would be good, especially when the basic math has always said that single-poll stories are irresponsible. (There would still be plenty of polls from universities and marketing firms.)

A case study of the perils of one-poll stories occurred in recent weeks in the race for governor of New York. Every one of the seven nonpartisan polls included in the RealClearPolitics average since late September showed Gov. Kathy Hochul with a lead of four to nine points; she ended up winning by about six points. But Trafalgar, a Republican pollster, issued two polls saying the race was much closer, the last of them claiming it was tied. The local press seized on each of the Trafalgar polls in turn, and that drove a narrative that should be embarrassing to every news organization that amplified it.

Some encouraging signs on the democracy beat

In one of the first of these columns, more than 19 months ago, I wrote that responsible journalism isn’t inherently progressive, but it is pro-democracy. So I have been gratified as leading newsrooms launched democracy beats over the last two years. I think one clear message we can and should read in the midterms is that this work has both responded to and struck a chord with voters.

Turnout was robust (if down from 2018), election deniers were mostly routed, particularly in seeking governorships and secretary of state posts in what have recently been swing states, while the AP’s modern equivalent of an exit poll showed that “the future of democracy in this country” was a leading issue for fully 44% of voters, second only to inflation.

It will be important to sustain attention when election deniers continue to seek power through elections (although it may start dawning on candidates that this is a difficult sale).

Two other issues on the democracy beat are worthy of more, and more rigorous journalistic attention. First is attempted voter suppression, which seems to me to continue to fail, except at the margins, but which seems also to be increasingly normalized. We need better stories, more quantitative and less anecdotal, after the votes have been cast, summing up the impact of these efforts.

Second is gerrymandering, which has become thoroughly bipartisan, with a lot of commentary almost commending actors (like Ron DeSantis) who practice it most ruthlessly and ridiculing those who seem more conflicted (like New York State Democrats), rather than laying bare how gerrymandering undermines the bedrock idea of republican government that people choose their representatives rather than the other way around. The problem, of course, is that the Supreme Court has permitted what amounts to a free-for-all on gerrymandering. A robust democracy beat might seek to place this question more firmly on the congressional agenda— and to question the Court’s motivations more closely.

The last seven years have been an especially challenging and disappointing period for political reporting in America. The rise of a would-be authoritarian was initially exploited for ratings and clicks, and often chronicled as if it were a Roman circus. If we are to emerge from this dark period, coverage of politics in the United States needs to change, and may be beginning to do so. I hope we can remain on that path.

Second Rough Draft will return in two weeks. Happy Thanksgiving!

Leave a comment

Second Rough Draft

Early Returns on Political Coverage and the…