This is such a wonderful piece, and, realizing that it could take an epoch for me to craft a response worthy of it, I thought I’d just post responses to a few of the issues you pointed out.
Dealing with data that’s not all that “open”
Its statistics — which only cover the bare minimum of indicators, for only the top 10 ranked players, and not even to the level of individual match data — are locked in PDFs.
That’s super annoying, and it’s not unique to sports. This remains true for a number of “public” records, and various datasets of interest to journalists, which (*hooray*) also means that there are resources out there to make the process of “liberating” the data a bit easier. I recently came across Drawing by Numbers, a resource on data visualization for activists by the Tactical Technology Collective, and they happened to have a guide for this very scenario: “How to: Opening Open Data — Accessing data when it’s inaccessible”.
Data Science & Data Availability
[Stephanie Kolvachik] noted also that with the success of data science comes more opportunities for commercialization and, therefore, a closed-door policy on collected data for those unwilling to pay the entrance fee.
While there’s truth to this, I shudder at the notion of blaming data science (and, by proxy, data scientists) as the cause of this problem. I could be wrong (I haven’t actually done any causal inference on this), but if data keep spectators engaged and watching, liberating said data should become a mutually beneficial prospect for “publishers” (e.g. ESPN).
Accessibility & Barriers to Entry
[Charles Allen] told me that some of the top coaches he knows — with all this technology at their disposal — never use it, because they don’t have the time or support systems to acquire and deconstruct the data.
All I can say here is: we’re working on it. I’m not an UI/UX person, but it seems like data tools for the non-technically inclined are proliferating rapidly, and many of them are mentioned here on Medium.
What does a girl have to do to get some data around here?
Allison McCann’s (whose work I always admire) “Hey, Nate: There Is No ‘Rich Data’ In Women’s Sports”, and Sue Bird’s piece from The Players’ Tribune (“Analyze This”) are both pretty dead on.
And so, it’s with some trepidation that I face these questions from your post:
I have to ask: Are people just not interested? And why not?
First of all, there’s the guilt, which I totally fess up to. I know very little about the WNBA teams and players. I’ve learned more by following the highlights shared by Gabriella Levine on twitter, and love what Excelle Sports is doing for the coverage of female athletes in sports journalism.
The flip side to the guilt is a sort of adolescent indignation — it seems unfair to assume that because I’m a female who “does” sports statistics I should be doing this for women’s sports. Though I certainly don’t watch all of them, there are over 1,200 games played in a single NBA season. And, yes, that’s a thing that people who like analytics do — they watch sports, just like everybody else.
What’s more, it’s not like analysis is done in the vacuum of a single season. Good statistical modeling is a creative process; it draws on a knowledge of both what is and is not being captured “by the numbers.” Thus, I’m quickly intimidated by the idea that I would have to somehow assimilate 20 years worth of WNBA lore to do it any justice.
However, do I think the data should be there? Absolutely! In fact, rich data can draw me into a subject I might otherwise ignore (e.g. the weather).
