Coverage of the 2016 presidential campaign has taken more twists than any other U.S. election in recent memory.
But Slate’s planning to add one more.
The online magazine will be breaking with Election Day convention in November by providing real-time election projections by working with a company called VoteCastr. The company, which is based in Palo Alto, California, will examine turnout data to forecast winners in crucial swing states.
Of course, this effort flies in the face journalistic tradition. Conventional media wisdom says news organizations should be circumspect about who’s winning early on in the voting so as not to suppress voter turnout or misrepresent actual voting margins.
But Julia Turner, Slate’s editor, isn’t concerned about upending that tradition. In an email question-and-answer session with Poynter, she says the current approach to reporting on Election Day goes against a key tenet of journalism — providing the public with information. That exchange is published in its entirety below.
In a recent interview with The New York Times, you call the current Election Day reporting approach “ill-conceived and anti-journalistic.” Can you expound on that?
The role of journalists is to provide the public with information. But the media approach to Election Day contravenes this principle. Information gathered in exit polls during the course of the day remains embargoed until the polls close.
The theory is that publishing information about how a race is going might depress turnout among those yet to vote. But there is no good evidence supporting that theory, or suggesting that publishing such information has any consistent effect. Which means that the media are essentially colluding with each other to make voters less informed than they might be on a hunch.
Are you concerned this new approach could affect voter turnout in battleground states? Do you worry at all that voters might see real-time election returns and be discouraged from casting ballots, or feel complacent after seeing their candidate is up in the early voting?
Again, there is no sound research that suggests a link between publishing real-time information on Election Day and depression in voter turnout. Our democracy gives voters great power, and we should trust them to make smart decisions based on the information they have and what it means to them. I also think there are a few other reasons why sharing this information makes sense.
For one thing, the Election Day blackout seems illogical in an era when more than 30 states allow early voting; so our democracy already lets voters make their decisions with differing levels of information about the likely outcome of the race.
It’s also important to remember that throughout Election Day, voters are hit with marketing—ads, text blasts, emails, calls — from campaigns that are looking at live turnout tracking, and shaping their marketing based on that information. Publishing our data will help level the playing field, so that voters know as much as campaigns do.
It’s not unreasonable to assume that Slate might be opening a Pandora’s Box of sorts here. After all, once one news organization starts using real-time voting data, it seems ludicrous to imagine the AP or Bloomberg or Reuters waiting to call a state. They could get scooped! Do you think your decision will see a stampede of news organizations following suit?
Well, first I should note that we’re not planning to call any states. We’re planning to offer snapshots of how voter turnout is looking in key states, along with analysis about what those numbers suggest about how various states might go. We’ll leave the final adjudication of the race to the networks when the results are in. As for what other outlets might do in the future, I wouldn’t presume to speculate, but their current approach privileges caution over informing voters.
I do think there are two parts of our approach that may gain wider adoption: the decision to cover turnout itself as part of the story of Election Day, and the approach our partners at Votecastr will use in their analysis. They’ve modeled their approach on what campaigns do to predict results, rather than what media groups do with exit polls, which are not designed purely with projections in mind.
With more time comes more certainty where calling elections are concerned. The more data you have, the more confident you can be about a state’s ultimate electoral decision. Are you worried at all that this lack of traditional deliberation might lead to inaccurate results? Why or why not?
Again, we won’t be calling states — we’re going to leave that to the networks. But we trust the system developed by Votecastr and its team of experts from both sides of the aisle. They’ll be using the same methodology campaigns use to drive their decision-making on Election Day.
We’ll be clear about the methodology and its assumptions, but we think information that is good enough for candidates is good enough for voters.
If the data is comprehensive, why leave the business of officially calling states for traditional players like the AP or CNN? It would seem to me that Slate would be well-positioned to call states if it had a running tally of how things stand throughout the day.
Well for one thing, we’re only going to be publishing data from a few key swing states. But more broadly, the opportunity to beat out the major players on election night is not the appeal of this partnership with Votecaster.
What we want to do is challenge the way the media broadly thinks about covering the election as it progresses, and to let voters follow the election the way insiders do. The voters are the most powerful people in the country, and they should be well-informed.