Analysis Paralysis: Simplifying Advanced Sports Analytics for Fans
...but also, how to lose people along the way
If you are new around here, here’s why I started writing.
I went to the Boston Bruins game this past weekend thanks to my most committed fan and reader (my dad - thank you). It was a great game for Bruins fans, with the team scoring nine goals against its historical arch-rival Montreal Canadiens. I stayed in my seat for the intermissions and I noticed these advanced stats on the jumbotron. I took a photo of course, because it was one of the first times I have seen this sort of activation in venue (bravo Bruins / Rapid7).
As I looked back at my photos and videos from the night, I took a deeper dive, with my thoughts now evolving into this blog.
While typical counting stats (assists, goals, time on ice, plus/time) can be tracked through any sports app, advanced analytics can help bring the full story to life and into context. Looking at the left image, as a Bruins fan, I know that Charlie McAvoy is on the ice more than most defensemen, as he is on the Bruins power play1 and also brings the puck up the ice quite a bit. Lindholm is his line mate, while rounding out the top five, Marchand, Pastrnak, and Coyle were on the first line of forwards.
On the right, the hardest shots table puts into context just how fast that puck flies at the goalies. Zacha and Heinen are forwards, while Carlo, Shattenkirk, and McAvoy are defensemen.
But, are you lost? What does it all mean? Why does it matter?
Not everyone is going to care about advanced analytics (especially with a few beverages in their system on a Saturday night at TD Garden), but these are really cool insights that tell a unique story - but these can be developed further.
Here’s how I approach this and look at translating advanced analytics to the general public, whether that be in venue, on broadcast, mobile, or any other means. We will focus on the wider general public, but I’ll dive into the discussion with a coach/athlete on the next blog.
Define goals
Provide context visually
Keep it simple by comparison
Make it memorable
To do this, I’m going to use the examples from the game this past weekend, but there are so many advanced analytical metrics out there these days. There are also so many ways to approach these metrics, but I’ll keep it consistent and simple here to get us started.
Define Goals
This is the most simple, but sometimes overlooked task. I start with defining goals by asking questions to then outline some broad ideas for how to make this work.
What is the purpose of this project?
Who are we talking to?
Where will this live?
When will this be used?
What story are we trying to tell?
How will we gather these stats?
These questions are simple, to the point, and help define what might we are trying to do. Most importantly, it will help define the key performance indicators that are related to the project. For example, here’s what the answers could look like for the fastest shot example:
What is the purpose of this project? To showcase advanced analytics to fans as well as provide a new sponsorship offering
Who are we talking to? Fans in the arena
Where will this live? Jumbotron
When will this be used? During intermission
What story are we trying to tell? Specific to the statistic displayed, which athletes are performing in the top five of the category
Provide Context Visually
Charlie McAvoy may have skated 1.4 miles, but not everyone knows exactly what I explained in the above regarding his strategy. There is nothing wrong with the table being used, but what it lacks is context, like time on ice, or lengths of the ice to make it really stick.
A heatmap (like below) would likely be too small to show on a jumbotron, but could be displayed through a QR code or push notification outlining the advanced stats campaign in the app. While this one is showcasing goals locations, this type of visual could show fans exactly where McAvoy is skating on the ice to reach the mileage.
The same can be done for goals or shots. Again, on this visualization by Jared Flores, he allows the user to pick a specific shot and see the information behind it. Using our ‘hardest shot’ example from Saturday’s game, this type of chart could show the location of their shots to put them in context.
Keep it Simple by Comparison
Another option for this same statistics is to make it really simple by comparison to known locations or relatable concepts. How many lengths of the ice is 1.4 miles? How far could McAvoy run on the Charles River Esplanade? Conveniently, the Boston Common is 1.4 miles in length, so this comparison would work perfectly and resonate with the fans in the crowd. The main need here is to set a standard for comparison and maintain it through all the games.
To bring this to life via mobile or on broadcast, it’s essential to put it into context for the every day fan. What does it mean to skate 1.4 miles?
Showcasing visually how many times this has happened throughout the season through a known feature or event within the market
“McAvoy has skated XX miles total this season, the equivalent of ## Boston Marathons”
AI predictions of how much he might skate this season, based on past years and current season to date
Giving those fans who want it an opportunity to go deeper into the analytics
One example of this last point is NHL Edge, which allows for a fan to get an overview through deep dives into various hockey players, situations, and advanced analytics. Most of us are likely also familiar with the NFL’s Next Gen Stats, which offer similar contextual analyses.
Make it Memorable
The most important thing to do is build off the visuals and the comparisons to make it memorable and create a journey for fans. We should be leaving fans wanting more, even if a data point is complex and requires additional steps. These are some of the questions I like to ask to better understand how to really bring the concept home.
Is this normal for McAvoy? How does this compare to his time on ice? What is his average and max/minimum across the season? Maybe this jumbotron table is a cheat sheet, while fans try to respond to questions or trivia on by quickly ranking the players that have skated the most. How do we encourage fans to complete the quiz? Offer an incentive.
If at the arena, maybe a coupon for the player who has skated the most, or has the fastest shot? If at home via broadcast, perhaps a sweepstakes for a video meet and greet with the player? Or, could we create a streak focused activation as discussed last week to maximize the journey from game to game?
The main focus should be to make it fun and memorable - creating a journey for fans into their understanding of advanced analytics. If that isn’t happening, then what’s the point of showing the stats (beyond stat nerds like me)?
Skating This Home
This serves as a starter pack on making advanced analytics more digestible and approachable for fans. While I used hockey today, there are so many different ways to bring this to life in other sports. This is generally how I approach my thought process in bringing these stats to life. To hit it home:
Define goals to outline key details and objectives for the project
Provide context visually to make the statistics easy to digest
Keep it simple by comparison using different measuring tools
Make it memorable by extending the journey
Hope you all have a great week - I’ll be back here during the week of February 12th for the next blog. Thank you for reading and sharing - it means a lot.
for non-hockey fans: a +1/2/etc player(s) advantage over the opponent - the other team committed a penalty and sits in a glass box for multiple minutes (literally)