Launch of FiveThirtyEight: An interview with Lead Writer Carl Bialik

FiveThirtyEight is due to launch today, the website and brainchild of Nate Silver, Editor-in-Chief of its blog formerly at The New York Times who correctly predicted 49 out of 50 states in the US Presidential election in 2008 by using Bayesian statistics. Four years later, he correctly predicted the winner of all 50 states and the District of Columbia during the 2012 US Presidential Elections between President Barack Obama and Senator Mitt Romney.

Ever since last July’s announcement from Silver that he and his FiveThirtyEight blog would depart The New York Times and join ESPN, the world of statistics has eagerly anticipated its launch. During his speech at JSM 2013, I recall statisticians present started to to tweet him asking how they could get an interview at FiveThirtyEight (for an exclusive video interview with Silver himself during JSM 2013, please click here to register for free and view). Recruitment has taken place over the past few months and one of those who has successfully won a role is Carl Bialik, who was the creator and writer of the weekly Numbers Guy column for the Wall Street Journal and will now be FiveThirtyEight’s Lead Writer for News.

I met Mr Bialik at the Future of Statistical Sciences Workshop when he had just won the role about which he was very excited. Here he talks exclusively to Statistics Views about what the world can expect from FiveThirtyEight, about his own career and how he came to be involved.

1. Please could you tell us more about your educational background and how you came to recognise statistics as a discipline in the first place? What set you on the path to becoming a journalist writing about statistics?

I began down the path of being interested in statistics at a young age, analysing baseball statistics with my father to nominate players for awards, and hearing my mother correct our use of the word “statistics,” since we were not accounting for variance and uncertainty. I took to math in primary and secondary school and concentrated at university in math and physics, close cousins to statistics. I also worked in science research — at a biophysics laboratory in secondary school, then at a nuclear-physics lab during university — which helped me learn how to analyse data.

2. You were the creator and writer of the weekly Numbers Guy column for the Wall Street Journal for several years, which covered the use and misuse of numbers and statistics in the news and advocacy. What was it that you aimed for with each column?

The goal was to put numbers up to the same scrutiny that good journalism applies to words. Numbers, to journalists and readers who aren’t comfortable with them, can seem intimidating and infallible. I sought to demystify them and sometimes debunk their misuse, often holding accountable politicians or companies who twisted data to serve their needs.

3. What was the most bizarre misuse of statistics that you remember coming across?

It’s hard to pinpoint just one. What comes to mind is the human-resources consulting company that each year reported a figure on how much productivity U.S. companies were losing to the men’s college-basketball tournament. Each time, the calculation differed, with changing assumptions; each time it also made the false basic assumption that every minute spent following basketball in the workplace was a minute of work lost. And each year some media outlets reported the figure with the same gravity they’d have lent to a peer-reviewed scientific study.

        Carl Bialik

 

4. What kind of feedback did you receive from the WSJ audience? Did they include statisticians or members of the public?

WSJ readers are very interested in economic and financial data, so they responded well to the column, often providing story ideas or asking questions that sparked columns. Statisticians were valuable sources for the column and also often provided feedback — including those unhappy times when I came up short in attempting to explain statistical concepts to laypeople whilst preserving accuracy.

5. You are also the co-founder ofGelf Magazine. Please could you tell us more about this growing online publication?

I started Gelf with friends during a period when I was freelancing for the Journal. It was a rewarding and enlightening experience in trying to create our own publication and business. Some of our coverage was similar to my Numbers Guy work, in holding people to account for misleading the public — for instance, movie advertisers who misquoted critics’ reviews. Today Gelf is primarily a venue for interviews with authors and other people who speak at Gelf-hosted events in New York.

6. What led to your new role at FiveThirtyEight and what does your role entail?

I was a longtime watcher of Nate Silver’s work on baseball stats and, more recently, in election forecasting for the New York Times. I admired his successful predictions and his ability to clearly explain statistics, so I was eager to join the group of talented editors and writers he was assembling to expand into new coverage areas, including economics and science. My role is the lead writer for news, which will share common ground with my Numbers Guy column for the Journal — in critiquing others’ work — while also involving more of my own data analyses.

7. What are the major challenges that FiveThirtyEight faces in terms of statistical analysis and critique?

One of the challenges, always, is getting access to good data on the subjects we want to cover. Sometimes good data aren’t public, and sometimes public data aren’t good. We’ll also have to make our coverage accessible to the lay reader and, at the same time, meaningful to stats junkies and quants.

8. Are many statisticians employed with you? How does the whole team work together?

Here’s our still-growing staff list http://www.fivethirtyeight.com/2014/02/fivethirtyeight-hiring-update.html. Andrew Flowers, our quantitative editor who joined us from the Federal Reserve Bank of Atlanta, will be a great help in vetting our data work. Many other staff members have extensive experience using statistical analysis tools and work together to tackle data sets.

9. Which statistical techniques are most commonly used at FiveThirtyEight?

It’s early to say, as we’re just getting going. Do expect forecasts, and follow-ups on the accuracy of our forecasts.

10. Could you please tell us about any projects you’ve done at FiveThirtyEight or are currently working on?

I’m doing some writing about certain consumer electronics and how they work, and about probabilities in the sports world.

11. I gather from recent interviews with FiveThirtyEight founder Nate Silver that there will be no subscription fee. What are the main goals for the new FiveThirtyEight and the challenges you as a team expect to face in the journey?

The goals are to broaden FiveThirtyEight’s coverage beyond politics into sports, science, lifestyle and economics, and thereby broaden readership of data journalism. We think our brand of journalism can provide a deeper and clearer look into just about anything that can be measured.

12. Please could you tell us more about the content plan for FiveThirtyEight? Are there plans to expand into podcasts, videos, webinars?

One of our features editors, Lisa Chow, formerly of National Public Radio, will serve as podcast host for us, so you can count on hearing a FiveThirtyEight podcast. We are exploring ways to tell FiveThirtyEight stories on television and in web videos as well. And of course data visualisation will be a large part of our work. Expect FiveThirtyEight to become increasingly visual as time goes on.

13. At the recent Future of Statistical Sciences workshop that we both attended, there was much talk about Big Data and a concern that many ‘hot areas’ such as big data/data analytics, which have close connections with statistics and the statistical sciences, are being monopolised by computer scientists and/or engineers. What do you think statisticians need to do to ensure their work and their profession get noticed?

Reach out to journalists to work on making their stories better and helping them to write ably about statistics. And do their own writing for a mass audience about issues in data that touch on people’s lives.

The goals are to broaden FiveThirtyEight’s coverage beyond politics into sports, science, lifestyle and economics, and thereby broaden readership of data journalism. We think our brand of journalism can provide a deeper and clearer look into just about anything that can be measured.

14. Is there a particular piece of work (research or otherwise) that you are proudest of?

I’m proudest of my body of work for the Numbers Guy column and accompanying blog. It had ups and downs, of course, but overall I think I played a modest role in increasing the public’s numeracy.

15. What has been the most interesting point or aspect that you have since learned on statistics or the portrayal of statistics?

It wasn’t surprising to learn that sponsored studies — often no more than public-relations exercises — are often flawed. Yet even peer-reviewed scientific work is subject to criticism; two statisticians might approach the same data entirely differently. Findings that so much scientific research isn’t reproducible are eye-opening, and cast doubt on how many researchers use the concept of statistical significance,

16. Finally, are there people or events that have been influential in your career?

Many. Each of my parents taught me about data and writing, as did many teachers and fellow students. Bill Grueskin, while editor of WSJ.com, had the idea for the Numbers Guy column, and many other Wall Street Journal colleagues helped develop it and guide me. And I’ve already got lots of ideas and motivation from my versatile and brilliant new colleagues at FiveThirtyEight.

 

Copyright: Logo appears courtesy of FiveThirtyEight and photograph appears courtesy of Mr Bialik.