Production line
Production line
Is UConn's influence on WNBA basketball waning?
Welcome to the fourth edition of Plot the Ball — a newsletter where I offer data-driven answers to interesting questions I have about the world of sport.
This month(-and-a-half) — after watching UConn’s unsuccessful run to the title game of the 2022 NCAA women's basketball tournament — I’ve been thinking about the impact that the Huskies’ college program has had on the WNBA, and wondering whether a run of six years without a national championship has led to its diminishing.
Is UConn’s influence on WNBA basketball waning?
The University of Connecticut’s defeat to South Carolina in the title game of the 2022 NCAA Women’s Basketball National Championship made it six years without a title for the sport’s flagship college program.
UConn’s current head coach Geno Auriemma has been in charge since 1985, and under his guidance the university have won 11 national championships in total.
That first win in 1995 also preceded the launch of a professional women’s league in the USA by a couple of years — and, over the last couple of decades, UConn has provided the WNBA with some of its biggest and brightest stars.
However, this recent stretch — following a run of four consecutive NCAA tournament wins between 2013 and 2016 — is the longest drought the program has experienced since UConn triumphed for the first time in 1995.
It is therefore worth stopping and asking: is this six-year spell an indicator that Auriemma’s program is no longer one of professional basketball’s elite talent factories?
First of all, however, it’s necessary to establish just how much of an influence UConn has had on the WNBA in its 25-year life so far.
Using data from Basketball Reference — and ignoring any minutes played so far in the ongoing 2022 season, which began on May 6th — we can see that no US college program’s attendees have played more regular-season minutes in league history than UConn’s.
But how have these minutes been distributed over time? Is the university’s level of influence truly on a downward trajectory?
That same Basketball Reference data shows that — for much of the league’s first two decades of existence — former UConn players accounted for a steadily increasing proportion of regular-season WNBA minutes.
In the inaugural 1997 season, the legendary Rebecca Lobo — a key part of that first national-title-winning team in 1995 — was the only UConn player in the league.
By 2018, however, a number of her fellow alumnae were among the WNBA’s most prominent players — including four past or future league MVPs — and many others were solid contributors to their teams.
Of the 124 players who played more than 200 regular-season minutes in 2018, 16 were UConn products — and approximately one in every seven minutes on a WNBA court was played by an athlete who had attended the university under Auriemma.
Since that point in time, there has indeed been a slight decline — not least because two-time WNBA champion Renee Montgomery (who played over 900 minutes in each of the 2018 and 2019 seasons) has since retired, and former MVP Maya Moore has stepped away from professional basketball after eight seasons to focus on her criminal justice activism.
However, things have certainly not dropped off precipitously: UConn products still accounted for 11.9% of all regular-season court time in 2021, equivalent to approximately one in every eight minutes played.
One interesting factor in this trend is the markedly different rules around when players become eligible to be drafted into the WNBA and its men’s equivalent, the NBA.
Following UConn’s run to the 2021 Final Four behind the strong play of point guard Paige Bueckers, Cassandra Negley of Bleacher Report summarised them well: “[m]ost college players aren't eligible for the WNBA draft until they've finished four years of college”, but “[a] player who turns 22 in the calendar year of the draft can renounce their NCAA eligibility and enter”.
In the NBA, by contrast, American players can enter the league after only one year of college attendance.
This works both ways. On the one hand, players like Bueckers and Azzi Fudd — blue-chip recruits out of high school, and likely legitimate WNBA prospects — are kept from competing at the highest level for longer than they otherwise might be.
On the other, the players who have featured during UConn’s recent run of leaner years would only recently have become eligible to play in the WNBA — leading to a natural lag between college success (or lack thereof) and the consequences of that you would expect to see at the next level up.
(Olivia Nelson-Ododa, for example, played in four of those title-less years at UConn, but has only just been drafted by the Los Angeles Sparks.)
But it’s also worth bearing in mind that competition for WNBA roster spots (and minutes) is incredibly intense. While there are 30 NBA franchises currently active, there have never been more than 16 in the WNBA — and, since 2010, that number has been 12.
Making it as a regular is therefore incredibly difficult, as Alex Azzi recently explained in an article for NBC Sports:
With all of that considered, UConn still accounting for more than 10% of all minutes played in a league that attracts players from all over the world remains mightily impressive.
What’s more, in the likes of Bueckers and Fudd, there might just be a few more WNBA stars on the way — as well as NCAA titles.
Further reading
Howard Megdal and Benjamin Morris of FiveThirty Eight on UConn’s run to the 2022 NCAA tournament and current WNBA superstar Breanna Stewart’s incredible college career
Jackie Powell of Bleacher Report on the aftermath of the team’s win over North Carolina State in the 2022 Elite Eight
Cassandra Negley of Yahoo! Sports on the WNBA Draft’s current eligibility requirements
Alex Azzi of NBC Sports on the reality which many players face after having been drafted
Technical notes
You can find the code for this piece on GitHub here
The major challenge I encountered in doing the data work for this edition was having to link two different datasets with common variables. As the standard Basketball Reference WNBA season statistics page did not include details of which colleges each player attended, I had to get a bit more creative: as well as scraping the basic table of statistics for each season, I had to write a script to grab the URL of each player’s individual profile — which did contain this information — too. And once I had all those links — each assigned to the name of a player in the main dataset — there was another tricky part. The basic information at the top of each profile was not set out in tabular format, so I had to wade through each text string and isolate the bit I was interested in. I was able to make this work, but it’s a section of the code I’ll definitely go back and review, as I’m already thinking about ways it could have been more efficient.
One of my instinctive reactions when it comes to visualising part-to-whole relationships (as in the second chart above) is that you should try and show the whole of the 0-100% range where possible — given that the existence of all of that other data (in this case, the share of WNBA minutes played by attendees of other colleges) is effectively implied, whether or not it is the focus. However, I acknowledge that this isn’t always practical — especially when what you’re focusing on is a fairly small proportion, you can end up losing the actual message you’re trying to convey. In this instance, I moderated somewhat and went for a 0-45% scale; I felt this gave a better sense of the general part-to-whole relationship than having the data take up most of the chart, and it also left me with room to add additional useful context through annotation.
I always wanted to think a bit more concertedly about how to structure the data-led narrative in this edition, having read Elevate Dataviz’s recent blog on the subject of ‘Data Stories’, and intended to structure this piece as a fairly straightforward ‘question story’. Hopefully all of the elements are clear enough in the text above:
“Question story: Answers a question with data using question -> hypothesis -> experiment -> results -> conclusion. Can include a Scale story to compare the size of one thing to another.”
Finally, I did all of the scraping for this manually as I couldn’t find a nifty ready-built R package to extract WNBA data from the Basketball Reference site. If you know of one, please pass on a link!
Next month — to coincide with the beginning of the 2022 State of Origin series in Australia — I’ll be looking at the on-ball work of the most creative players in the recent history of the world’s leading club rugby league competition.