- The FiveThirtyEight archive preserves nearly 15,000 articles from one of data journalism‘s most influential outlets.
- The FiveThirtyEight archive shows Nate Silver alone authored nearly 5,000 pieces — a staggering editorial output.
- Disney’s quiet shutdown of FiveThirtyEight in 2023 erased a decade of pioneering statistical reporting from the open web.
- A fan-built index now makes it possible to browse the entire catalogue by year and byline for the first time.
- The FiveThirtyEight archive preserves nearly 15,000 articles from one of data journalism’s most influential outlets.
- The FiveThirtyEight archive shows Nate Silver alone authored nearly 5,000 pieces — a staggering editorial output.
- Disney’s quiet shutdown of FiveThirtyEight in 2023 erased a decade of pioneering statistical reporting from the open web.
- A fan-built index now makes it possible to browse the entire catalogue by year and byline for the first time.
The FiveThirtyEight Archive Is Back — And It’s Bigger Than You Remember
The FiveThirtyEight archive now lives on the Internet Archive, and someone has gone to the trouble of building a proper index for it. The site, fivethirtyeightindex.com, lets you browse thousands of articles by year or by author — and the sheer scale of what’s there is a little breathtaking. This wasn’t just a blog. It was a publishing operation that churned out serious, data-driven journalism for over a decade, and the numbers behind it tell a story that’s easy to forget now that the site is gone.
FiveThirtyEight launched in 2008 as Nate Silver’s personal project — a place to apply baseball-style statistical modelling to political forecasting. It called its shot spectacularly in that year’s presidential election, correctly predicting 49 of 50 states. The New York Times picked it up in 2010, ESPN absorbed it in 2013, and ABC News took over in 2018 after Disney consolidated its media properties. Then, in mid-2023, Disney pulled the plug. The editorial staff was laid off, the site went dark, and years of original work vanished behind a dead URL.
That’s what makes this index worth paying attention to. It’s not just a nostalgia project. It’s a record of what a certain kind of ambitious, data-forward journalism actually looked like — and how much of it there was.
The Scale of What Was Built
Look at the byline counts and the numbers are genuinely hard to process. Nate Silver himself is credited with 4,966 articles. That’s not a typo. Nearly five thousand pieces from the outlet’s founder and most prominent voice. For context, that’s more than one article per day, every day, for over thirteen years. Even accounting for short-form posts, quick model updates, and liveblog entries, it represents an output that most journalists wouldn’t approach across an entire career.
Neil Paine comes in second with 1,428 articles, followed by Walt Hickey at 1,210 and Aaron Bycoffe at 1,168. The sports analytics specialist Josh Hermsmeyer contributed 122 pieces. Rob Arthur, who did some of the most interesting work on baseball pitch data and player safety, clocked 120. These aren’t hobbyists. These are people who built genuine expertise in a specific intersection of journalism and statistical analysis — and did it at scale, in public, under deadline pressure.
The full list runs to dozens of writers. Maggie Koerth, who covered science, put in 319 articles. Mona Chalabi, who went on to win a Pulitzer Prize for her illustrated data journalism at The Guardian, contributed 242 pieces during her time at the outlet. Clare Malone, now at The New Yorker, has 360. Harry Enten, who’s become a familiar face on CNN, contributed 671. The FiveThirtyEight archive is, among other things, an early-career portfolio for some of the best journalists working today.
Why Data Journalism Lost Its Flagship
It’s worth being honest about FiveThirtyEight’s complicated legacy. The site had its critics, and not without reason. There were genuine debates about whether probabilistic forecasting — publishing a model that says Candidate A has a 71% chance of winning — was more confusing than clarifying for general audiences. After 2016, when Silver’s model gave Hillary Clinton roughly a 70% chance of winning the presidency and Donald Trump won anyway, the backlash was fierce. Never mind that a 30% probability is not zero, and that the model’s uncertainty intervals were actually wider than most competitors’. The perception problem stuck.
The site also wrestled with internal tensions that became public. In 2022, staff voted to unionize. In 2023, before a contract was finalized, Disney announced it was shutting the operation down. The timing was pointed. Whether or not the two things were directly connected, the optics were bad, and the journalism community noticed.
But strip away the politics of the ending, and what the FiveThirtyEight archive represents is something that’s genuinely hard to recreate: a sustained, well-resourced editorial environment where quantitative methods were treated as a first-class journalistic tool rather than a sidebar or a gimmick. That combination — editorial ambition, data infrastructure, and an audience willing to engage with uncertainty — doesn’t come together often.
What the Internet Archive Preservation Means
The Internet Archive has become an increasingly critical institution as commercial publishers shut down, restructure, or quietly delete older content. FiveThirtyEight is a high-profile example, but it’s part of a much broader pattern. Gawker. The Awl. Deadspin’s archives after its ownership change. Countless Tumblr blogs. The web is not as permanent as it seems, and large swaths of the last two decades of digital publishing have simply disappeared.
What the fivethirtyeightindex.com project does is make the archived material actually navigable. The Internet Archive’s Wayback Machine is powerful, but it’s not designed for editorial browsing. You need a URL to start with. The index solves that by providing structured access by byline and date — turning a raw archive into something closer to a usable library.
It’s a volunteer effort, which is both admirable and a little depressing. The fact that preserving this body of work required a third party to step in, build an index, and host it independently says something uncomfortable about how the industry treats its own history.
FiveThirtyEight Archive as a Blueprint — and a Warning
The data journalism space didn’t die with FiveThirtyEight. The Pudding continues to produce remarkable visual essays. Reuters Graphics and the New York Times graphics desk do work that would have felt like science fiction a decade ago. ProPublica’s data team has broken major investigative stories using computational methods. The approach has dispersed across the industry rather than concentrating in one place.
But the concentration was part of what made FiveThirtyEight distinctive. Having a critical mass of writers who all shared a statistical worldview created a kind of editorial culture that was self-reinforcing. Writers pushed each other. Methodologies were debated openly. Mistakes were acknowledged in print. That culture produced work like Benjamin Morris’s deep dives into NBA player aging curves, or Rob Arthur’s analysis of spin rate manipulation in baseball — pieces that wouldn’t have existed anywhere else at the time.
The FiveThirtyEight archive is a record of what that culture produced over roughly fifteen years. Whether anything quite like it gets built again depends on whether someone is willing to fund it — and whether the economic model for serious, data-driven digital journalism ever stabilizes enough to support it. Right now, neither of those things looks especially likely. Which makes the archive more valuable, not less.
