How did your passion for sports begin, and where do your sports loyalties lie?

I’ve followed sports for as long as I can remember. My dad was a high school football coach in Iowa, and he’d come home from games on Friday nights with stats books. Around 4th or 5th grade, I started calculating stats for him. By junior high and even before I could play high school football, I would go through old stats books and do tackle charts.

I was also in a lot of crazy fantasy leagues. I started my own league in high school and calculated most of the stats by hand, then created complicated draft models and collected data so I could keep building them.

When I was in college, I applied for and was invited to play in an All-Star fantasy baseball league which was very in-depth. For example, in addition to a full major league roster we also drafted a half dozen minor leaguers. So I created models based on the players’ age and performance to find the best prospects. I’d scrape stats, then use Excel and its macros to make my own models and generate reports. Those reports did pretty well; I identified Bobby Abreu, Vladimir Guerrero, and Andruw Jones and got them on my team, which gave me some long-term success.

Soccer is now one of my favorite sports right there with baseball. That's a relatively new development as I wasn't previously aware of professional soccer beyond the World Cup. We began plans to launch FBref in 2009 and as part of that project, I started watching club soccer on ESPN, Fox Soccer and elsewhere. I was hooked after that. I overcame a pretty severe Football Manager addiction in 2013 (4 UCL titles with Hertha Berlin, thank you very much), but now I enjoy club football a great deal and I'm an active USMNT supporter as well.

Why did you decide to pursue an undergraduate degree in Mathematics?

Math was always my favorite subject. I’ve enjoyed it for as long as I can remember, and I had been accelerated a few grades ahead in math as well. When I majored in Mathematics at Grinnell College, I didn’t really know what I could do with the degree. I knew I didn’t want to be a high school teacher, but I grew up in a very small town and there weren’t many examples of mathematicians working there, so I thought I could be an engineer.

I went to engineering school for two years at Iowa State while finishing my degree at Grinnell, and then I ended up not liking engineering, so I returned to graduate school to pursue and complete a PhD in Math Sciences. So, I went from Grinnell to Iowa State, then to the University of Iowa.

What advice would you give to someone looking to pursue a career in sports, or enter the sports industry?

My primary advice is that no one is actively stopping you from writing, thinking about or doing things in sports. I get a lot of requests from people who want to do analytics for teams, so most of what I write below relates to that, but it is similarly true for other areas. For instance, Jon Bois didn't wait for permission to start writing and creating fun and interesting content around sports. His work created its own audience and led to opportunities for him with SB Nation and elsewhere.

If you want to do analytics for a team, start doing analytics for yourself and for a public audience. You can publish a great deal online, perhaps first on social media, Reddit, or Discords of like minded folks. If you do a bit of hunting, you can probably find places to share your work. You might ask people you enjoy reading if they have places where you can publish work. Getting feedback from peers is the fastest way to improve your work and find the gaps in your skills and thinking.

I suggest you try to create models or studies that find interesting results of a small scope. Tell me something about how shot selection for centers has changed rather than creating an all in one value metric for the NBA. No one will believe that you have a better all in one metric than a team can create for themself, but you might pick apart and study one small part of the game and actually demonstrate something new to people who have been studying the sport for a long time. A first place to start might be to think about what is conventional wisdom and then check if that wisdom is actually correct.

In terms of skills to have, I think that skills around data acquisition and cleaning are very useful. A good data analyst will often have to deal with incomplete or hard to access datasets, so the ability to gather your own data and wrangle it into a useful format will be a valuable skill. Then, I would say skills for analyzing and summarizing data would be next. This could be in SQL, R or Python or something similar, though SQL, Python and R are certainly current industry standards.

Lastly, you must work on your ability to communicate. A groundbreaking result poorly conveyed has very little value. What you’re describing and why it’s important must be clear to your audience. This audience is likely busy and will look for many reasons not to read your study. Take your writing and verbal communication skills seriously and work hard on them. If you are still in school, a technical writing course may be useful (if it's done well).

Why did you start Baseball Reference? What were your hopes and aspirations when you created it?

I published player rankings online on a site called iowafarmreport.com and that gave me some notoriety, and led to an invitation to write a book (Big Bad Baseball Annual 2000). When I was in graduate school, I spent some time running their website and thought having an encyclopedia would draw people to the site. In early 2000, I launched a site called Big Bad Baseball Stats (bigbadbaseball.com/stats/) and that quickly became popular.

Within a month or two, this site was becoming very popular, so I decided to spin it off into its own site, Baseball-Reference.com. I've come to terms with it, but for a long while, I was upset I had chosen such a generic name. I filled notebook pages with new possible names and considered changing the name as recently as 2012.

I started with player, team, leaders and awards pages. The first set of stats I used on the site came from Sean Lahman's Baseball Archive database, which was a free resource on the web. It was based on the data in Total Baseball 4 and the CD-rom that came with the book. I was living in Macon, GA at the time. My wife was a new math professor at Mercer University and I was in graduate school trying to avoid completing my dissertation. It took me about 2-3 months to create the basic pages and they've been the same format since then.

I knew that hyperlinks would be tremendously helpful for working with this baseball data. The giant books that held this data earlier were difficult to navigate and being able to jump from Barry Bonds page to the 1988 Pirates page to Andy Van Slyke's page with a couple of clicks would be a massive upgrade. Of course, no one was imagining that we'd also be able to do all of this on a handheld device, so initially it was just for desktop, and given that many people were still on dial up modems, I was laser focused on making the pages as fast as possible. We've had that aggressively functional aesthetic since then.

I did that site as a side project for six years. Along the way, Doug Drinen launched Football Reference in December of 2000 and Justin Kubatko launched Basketball Reference in 2004. The sites were co-hosted on the same servers and shared some infrastructure from the start.

How has working in data changed from when you first started Baseball Reference?

When I started in the late 1990s and early 2000s, there was very little work being done inside team front offices. There were random analysts here or there, but mostly teams did not do sabermetrics and the league certainly didn't do much data tracking at all.

The expansion in these areas started around 2000-2003 with the hiring of Bill James and Bob Bowman at MLB Advanced Media. MLB quickly started collecting data and moved that in-house rather than relying solely on the Elias Sports Bureau. After the success of the Red Sox in 2003 and 2004, teams started hiring analysts and haven't stopped.

Advances like PitchFx (in 2007-ish) and Statcast have just accelerated the trend, so that now all teams have fairly sophisticated data ingestion and analytics capabilities, though they're far from done studying this data.

It makes me a little sad that we went from a world where all of the interesting work was happening on our site, or Baseball Prospectus or Fangraphs, to now, where the teams are hoarding knowledge and the best data is difficult to come by and not released to the public.

Interested in hearing more about Sean? We’ll be updating this page soon with additional questions!