The New York Times' new data journalism site The Upshot launched on Tuesday, making it the third big new media site to focus on explaining the news with lots of numbers, following Nate Silver's FiveThirtyEight and Ezra Klein's Vox.
The Upshot will focus on politics and economics, and it has two goals, editor David Leonhardt writes in his intro. It will "navigate the news," for one, explaining what matters in a "direct, plain-spoken way, the same voice we might use when writing an email to a friend." In addition, the site will actively use large data sets to write those explainers. For example, its lead story today is a statistical model that attempts to forecast who will win a majority in the Senate in the 2014 elections. And there's also a nifty graphic that uses the model to simulate the election, given those forecasts.
The Upshot has only been up a few hours, sure, but it's impossible not to compare the site to the fellow fresh-faces on FiveThirtyEight and Vox. The three sites all live in the realm of what Al-Jazeera America's Malcolm Harris calls "Actually Journalism," or stories that cut through the noise to find that "Actually, X was really Y the whole time." You can see elements of that Actually Journalism already on The Upshot, including in "Incumbent Southern Democrats Are Less Vulnerable Than You Might Think" and "The Myth of Swing Voters in Midterm Elections." You think it's one way, the articles argue, but actually it's another.
Put those three sites on a scale of big data to explainer, though, and you can start to see the differences from FiveThirtyEight to The Upshot to Vox, respectively.
FiveThirtyEight rests on the far data side of the scale, focusing on data and regression analysis for sports, politics, and culture. Its story on the large number of deaths among professional wrestlers, for example, looked at a mix of actuarial and probabilistic death rates compared to their actual deaths. That heavy use of data comes to the conclusion — yes, wrestlers die more — but the explainer part gets less emphasis. The highlighted 0.0001 probability in the lower right of the chart means that the data on increased death rates of wrestlers is significant. But that conclusion isn't broken down into explainer-speak and written on a conversational level.
Contrast that with Vox, which resides on the far explainer side of the data-explainer scale, focusing more on writing in clear language for the common man. Vox's Cards in particular try to explain an entire topic in a few charts or paragraphs. When the site launched two weeks ago, The Wire's Allie Jones noted its similarity to Wikipedia or About.com pages for a range of in-the-news topics.
The explaining aspect has been somewhat helpful for basic topics, but Vox has so far had less success with its use of data. Matt Novak at Paleofuture slammed Vox's take on increasing adoption rates of new technologies, and a Vox story that posited Kansas led the nation in pornography watching was actually a misreading of data that defaulted its location to the center of the United States. It's only been a few weeks, of course, but Vox's strength seems to lie more in its ability to explain ideas than to digest complex data.
And so we come to The Upshot, which seems like it would like to position itself in the middle of the two ends of data and explanatory journalism. It does have one advantage in a talented graphics crew that, at this early stage, has done a good job breaking down complex data into interesting charts. See, for example, the color-coded chart below on the high reelection rates for Southern Democrats.
So yes, it's far too early to debate which of the three sites is "winning." And given that they have all positioned themselves in varying spots along the data-explainer scale, there's a good chance all three can live in their own particular field.