Nate Silver's FiveThirtyEight is here: may the odds be ever in its favour

http://www.theguardian.com/world/2014/mar/18/nate-silver-fivethirtyeight-odds-ever-in-its-favor

Version 0 of 1.

The latest of the new-wave of journalism websites is live. It’s FiveThirtyEight.com, the new ESPN-owned vehicle led by Nate Silver, formerly of the New York Times.

Silver, who built his reputation first with baseball predictions then with the best US election forecasts in the business, has turned his hand to launching a standalone data journalism website – but is it any good?

The first impression is that the new journalism looks a lot like the old journalism, though done well. No one who’s used a web browser for more than five minutes will find FiveThirtyEight difficult to navigate: the site’s eclectic selection of content (politics, sports, science, economics and “life”) is clearly delineated and color-coded, while the content looks familiar to anyone who’s looked at Wonkblog, the original FiveThirtyEight, or the Guardian’s Datablog: a headline followed by a few hundred words, interspersed with the odd graph.

The content is both good and varied, though. Day one reports included a lengthy post on how statisticians could help find the missing Malaysian airplane, to the accuracy or otherwise of the pedometers beloved of the quantified self movement, to a debunking of the old US myth about British teeth (a particular favourite of this British reviewer).

The writing is accessible, though tends towards the lengthy, and is willing to demand a certain level of familiarity with figures and complexity from its audience – not surprising, perhaps, for a site with a formal focus on data journalism. Posts tend to go beyond what’s typical in journalism, with some featuring original statistical analysis, error bars on (some) graphs, and a desire to keep firm research at the core of even the lighter pieces.

One of the best, though underrated, things about the FiveThirtyEight site at the New York Times was its presentation of data through graphics, requiring a largely unheralded team of interactive designers to present the detailed information, plus people to ensure the automated feeds of new polling info was fed in.

The early signs of the new standalone site on this front are good. Perhaps in a bid to move away from its perceived image as a polling site, there’s no major elections graphic yet. But the interactive graphic that is there – on the NCAA college basketball tournament – is clear, informative, and nicely presented. Given the data available from ESPN, and their expertise in this area, this should be a continued strong point for the site.

The new venture is going to face some challenges, though. The first is a simple one: its writers have had weeks, if not months, to prepare content for day one. They may also have some pre-cooked content for the next few weeks. There’ll come a time, though, when that reserve is depleted, and the ideas must flow from the week’s events, research and claims. Maintaining a quality start is a challenge, but one the site’s team should be able to manage.

Keeping an audience may prove more difficult: FiveThirtyEight faces a lot of competition. While it may be the only standalone data journalism website, it’s competing with sections in numerous other newspapers: The Upshot, in the New York Times; Wonkblog in the Washington Post; and the Guardian’s own data blog (now five years old) to name just a handful.

Rival startups, such as Ezra Klein’s explanatory journalism site Vox, plus statistical and data journalism from other upstarts like Buzzfeed are also muscling in. FiveThirtyEight isn’t focusing on one topic or issue, or a redefined way of presenting information: it’s trading on Silver’s brand, and a hope that people will buy into its data-driven approach.

That could work, though some may be put off by the in-references: the site’s name is a now somewhat dated reference to the size of the US electoral college for presidential elections, and bears little relevance to the broader purpose of the new venture. Its mascot, a fox, is a reference to a quote beloved of stats nerds and commonly adopted by Silver. Both may confuse new visitors.

The US has a population of around 300 million, only 96,000 or so of whom are statisticians (according to US Labor statistics). For sites like the Datablog, most of the audience is less interested in data per se (though a core is), but rather the topic the data is used to explain. The site will, consequently, likely rise or fall based on how well its writers come to dominate their verticals.

A final question mark as to that goal perhaps comes from the diversity of otherwise of the site’s masthead. Photo profiles of seven writers are featured on the site’s front page, including two talented Guardian alums: Mona Chalabi and Harry Enten. Six of the seven are men – and young, somewhat nerdy men at that.

While there is nothing wrong with any of those things – it’s a description that could easily be applied to your reviewer – the diversity or otherwise of FiveThirtyEight and similar startups could shape their future: can they appeal to a broad enough spectrum of gender, age and ethnicity? Do they even want to? Emily Bell, of the Tow Center for Digital Journalism, has discussed this at greater length here.

FiveThirtyEight is a data journalism website done well: it’s not reinventing the trade, but it’s markedly ahead of almost everyone else on quality, has a clean design, and a strong team. The site’s future will depend on whether that’s distinctive enough a proposition, whether the quality can be maintained, and how well it faces off the inevitable competition. But for now at least, it’s probably worth adding to bookmarks.