Why Nate Silver can’t explain it all

Between media startups Nate Silver’s FiveThirtyEight.com and Ezra Klein’s Vox.com, the newsplaining corner of the online media business is about to get a lot more competitive. Not quite reporting and not quite opinion writing, this data-driven journalism doesn’t simply strive to contribute to public policy and culture debates, it aims to end them with a decisive answer. Both sites assume a full understanding of the broad range of subjects they’re covering, and sufficient information to adjudicate all questions with responses that could invariably begin, “Actually …”

Welcome to the age of Actually Journalism.

Launched to coincide with the March Madness NCAA basketball tournament, FiveThirtyEight is the glasses-wearing baby of Silver, the statistics whiz who left a high-profile spot at The New York Times to develop the site with ESPN. In his manifesto “What the fox knows,” Silver explains that the new site will spread itself across five categories: politics, economics, science, life and sports. Considering the rigor Silver is associated with, the breadth of topics seems ambitious. FiveThirtyEight’s mascot and logo is a fox, a reference to the classic philosophical metaphor of the hedgehog (which knows one big thing) and the fox (which knows many things). Though ESPN has operated the long-form site Grantland at a loss for a couple of years now, FiveThirtyEight’s first round of content suggests Silver is going for something more clickable — closer to a Slate-by-numbers.

While FiveThirtyEight is the sugar cereal of this brave new content environment, Klein’s Vox comes complete with its own explanatory food metaphor. The result of another couple of highly publicized departures — Klein from The Washington Post’s Wonkblog and his buddy Matt Yglesias from Slate — Vox is the new flagship for Vox Media, which already owns successful sites such as SB Nation for sports and The Verge for science and tech. In an introductory post, “Nine questions about Vox,” the editors talk about the way journalism refers to articles on “hard topics” as “vegetables” — readers don’t want to eat them, but they’re good for you. However, they assure the reading public in proper 21st century foodie fashion, that vegetables “can also be roasted to perfection with a drizzle of olive oil and hint of sea salt.” Vox describes its mission as simple: explain the news. To the news they add “pop culture, science, business, food, sports, and everything else that matters.” Though Klein lost the Wonkblog name to the Post, the new venture is just as centered on the authors’ nerdy credibility and comprehensive understanding of the issues. Klein came up through the faux-meritocracy of policy blogging, delighting Democrats with his decidedly unsquishy analysis. Benjamin Wells, the author of a New York magazine profile of Klein (“Here, let Ezra explain”), describes his vision for Vox as “a bit like Wikipedia entries written by professional journalists.” (Who would ever aspire to such a thing?)

Vox has yet to launch — coming, like an unripe blockbuster, “soon” — but FiveThirtyEight’s early content has left a lot to be desired, even from other number geeks. Silver set the data bar high, promising in his manifesto to root out “incautious uses of statistics when they arise elsewhere in news coverage,” which is tough to balance with the recklessness required to compete in the online short-form market. So far the site has been a bit incautious itself — both economist Paul Krugman and Mother Jones’ Kevin Drum quickly found fault with FiveThirtyEight’s “chief economic [sic] writer” Ben Casselman’s dramatic use of revised data on corporate cash hoarding. When Casselman writes that a downward revision to estimates of total corporate cash holding undermines the whole narrative of corporations hoarding cash, he fails to consider carefully the basis for that narrative, let alone realize that he’s recklessly writing a new one.

When commentators, journalists or guys at parties presume to explain, they start from the presumption that they already understand both sides of the argument and have come to a definitive conclusion. There are times when this “Actually …” mode is not only appropriate but also enjoyable. When eminent astrophysicist Neil deGrasse Tyson rants on Twitter about the inaccuracies of space as depicted in the movie “Gravity,” everyone can laugh as a charismatic hedgehog pokes his spikes into pop culture. And though “Tyson Explains It All” would probably do well on premium cable, he sticks to his one big thing. When foxes start explaining, however, it’s time to check on the henhouse.

Nate Silver envies scientific objectivity, but he doesn’t pretend to practice it. In “What the fox knows,” he explains the magic trick of how a data point becomes a headline at FiveThirtyEight, a process he breaks down into a pseudoscientific method: “Collection, organization, explanation, generalization.” It’s no coincidence that interpretation isn’t on the list. FiveThirtyEight aims for the divine “truth beyond our perceptions.” In Actually Journalism, news and opinion aren’t needed; to understand something, all the audience needs is this fact, this piece of data, this answer.

If this all sounds a bit like a backlash to wishy-washy postmodernism, it probably should. With the rate of data accumulation accelerating to truly excessive levels, it seems as if we should have enough measures to extrapolate quantitative truths about behaviors we previously assumed could be evaluated only qualitatively. Pomo looks passé. Why shouldn’t reporting benefit from these same advances?

The late 20th century epistemological shift against privileged knowledge (as “The Big Lebowski” had it, “That’s just, like, your opinion, man”) was about more than destabilizing truths. With it came “standpoint epistemology” and the now maligned privilege check, which encouraged explainers to consider their own place in the story they’re telling, what they’re predisposed to focus on and what they might miss as a result. Checking your privilege has been obnoxiously caricatured as a way to silence whites, men, heterosexuals and other people on top in American social divisions rather than a method that might improve the quality of our thought. This knee-jerk ridicule means page views, a cavalcade of “_____ privilege” jokes and the same already tired commentators striking the same defensive pose. These new sites aren’t the only professional newsplainers — Gawker’s Hamilton Nolan and Slate’s William Saletan are other notable examples — but as eyes turn to Actually Journalism, its blind spots will become more apparent.

A glance at Silver’s archive at the Times yields a particularly egregious example: In December 2010 he set out to acquit WikiLeaks founder Julian Assange of sexual assault allegations in the court of statistics. Silver used a technique called Bayesian inference, in which he split factors into underlying conditions and new information, arriving after some analysis at the conclusion that the charges against Assange were (likely) trumped up. But one underlying factor Silver wasn’t considering was himself. A survey of research in the 2002 paper “Effects of victim sex and sexual orientation on perceptions of rape” in the journal Sex Roles cited a bundle of studies finding men “more likely to believe that a rape victim is responsible for her own victimization, that she provoked the assault by her relationship with the perpetrator, that she wanted sex, or that her behavior invited the rape.” Could an evaluation of his own underlying conditions have affected Silver’s calculation? What would the results of “critical” Bayesian inference look like? That’s the kind of truth that I fear will remain “beyond our perception.”

Another of Casselman’s first FiveThirtyEight posts, “Typical minimum-wage earners aren’t poor, but they’re not quite middle class,” is based on a single set of data plotting minimum-wage workers by household income. Casselman doesn’t consider the much-contested issue of how to measure poverty or the decrease in the real value of the minimum wage, and the only worker whose life he bothers imagining is the barista daughter of Fortune 500 executives. “How you interpret these numbers depends on your perspective,” Casselman writes, but the clear thrust of the piece is to undermine the case for raising the minimum wage by characterizing it as a rich kid’s problem.

Data extrapolation is a very impressive trick when performed with skill and grace, like ice sculpting or analytical philosophy, but it doesn’t come equipped with the humility we should demand from our writers. “We’re not sociopaths,” Silver has said, but that’s not something that someone who isn’t often accused of being a sociopath usually has to say. When confronted with the lack of racial and gender diversity on their staffs, Klein and Silver have hemmed and hawed about qualifications and blown it off, but if Actually Journalism can’t find a way to examine its own underlying conditions, it will be actually worthless.