MIT Sloan Sports Analytics Conference 2018 Recap

What is the future of analytics in sports?

James Anderson
Slalom Technology
9 min readFeb 27, 2018

--

Over the last 15 years in sports, sports analytics has centered on what I like to call the “Moneyball” approach to analytics. The popularity of the book by Michael Lewis gave life to the idea that analytics could drive how you built a team to compete, not only in the MLB, but also in the NBA, NHL, and NFL. However, as that type of analytics has become the standard, a new wave of sports analytics has swept through the industry. Last weekend, I attended the 12th Annual MIT Sloan Sports Analytics Conference here in Boston and heard from industry experts from every level about the future of sports analytics. What I found is that as technology has advanced significantly over the last 15 years. The application of analytics to sports has spread far and wide, touching many parts of the organization, and covering a wide range of use cases. I’m going to break them down into three categories, but ultimately, it boils down to the following need within the industry: near real-time information on every individual within the arena, both on and off the field.

Player Analytics

Over the last 15 years, many new statistics have been created, combining the classic statistics like home runs or Batting Average in baseball, completions and touchdowns in football, or points in hockey and basketball, to come up with new usage stats like Wins Above Replacement (WAR) in baseball or Quarterback Rating (QBR) in football. But with new technology comes new capabilities - and that fact has not been lost on the sports world. Many of the panels were about how player analytics can evolve to include concepts in machine learning, deep learning, and spatial recognition.

One very interesting presentation was from Raúl Peláez Blanco, Head of Sport Technology Innovation for FC Barcelona, who spoke about how they have revolutionized how they collect and analyze data. They use all sorts of tools to collect every piece of data they can on their players, including an ultra-wide band tracker that is strapped to the players during training. This tracker not only collects biometrics on the players, but also collects their location on the pitch. They’ve also built a platform that centralizes all this data (over 600,000,000 records) and runs deep learning on this data. This helps them run analytics against how positioning on the field affects the outcome of the match.

This same idea was also talked about at length in a hockey panel, which included John Chayka, General Manager of the Arizona Coyotes, and Dan Bylsma, former Stanley Cup winning Head Coach of the Pittsburgh Penguins. When asked about the future of analytics in the NHL, both John and Chris Snow, Director of Hockey Analytics for the Calgary Flames, agreed that being able to run spatial analysis on the ice in near real time to help measure a player’s “hockey sense” during the game. This level of deep learning and analytics is not only groundbreaking in sports analytics, but really in all types of analytics.

These new applications of technology was talked about at length in a panel called “Technology for a New Edge”, which included both athletes and technology industry pioneers. When asked about what technology they thought would be a total game changer, both industry experts (Terry Myerson from Microsoft and Adir Shiffman from Catapult) felt that the future was in wearable technology, and how that can feed AI to make coaching and player decisions. What was interesting though, was the 2 athletes on the panel (C.J. Anderson of the Denver Broncos and Chris Capuano who is a former MLB pitcher) were more interested in technology that allowed for game prep to be improved, like virtual reality for a pitcher to scout the next lineup he would be facing.

All of the panels, however, were cautious about data privacy and the impact of this data on players lives. Both C.J. Anderson and Chris Capuano raised concerns that this data could be used against them, whether by the coaching staff or management during negotiations. Chris Snow and John Chayka expressed concerns over how players and coaches react to this kind of data. They gave some examples, like Sidney Crosby, who thrives on getting all the information possible about himself, versus other players who react very negatively to this level of invasiveness. So, as sports analytics evolves, and more and more data is collected, the protection of the information must continue to be taken into account.

Fan Analytics

As technology has progressed, it has become much easier to get a ticket to any sporting (or entertainment) event. The secondary market boom, led by companies like StubHub and SeatGeek, has made it very easy for fans to buy and sell tickets to anyone who wants to attend. Rarely do fans go to the ticket window on game day and buy a ticket right then and there. Scalpers on the street have become less and less prevalent. Some organizations have even moved to a completely digital model, with all tickets being sent via a mobile application. But what does this mean for the organization? How can they leverage this for a better fan experience, while increasing their own attendance?

One of the biggest use cases that I see for a new analytics platform, industry aside, is the idea of trying to get a 360 degree view of a customer. In sports, the customer for every team are the fans, and many sessions focused on the analytics of the fans. The end goal for organizations, now that the technology is there, is to understand exactly who is in every seat in the arena. Historically, it was nearly impossible for the team to engage with their average fan. Their main goal was to satisfy season ticket holders so that they would continue to renew, providing them with a more personalized experience while letting the product on the field try to attract fans to fill the other half of the stadium. Yet, as tickets have become digital, finding a way of tracking the chain of custody of a ticket has become paramount to fan engagement.

In a session all around ticketing analytics, executives from teams and ticketing agencies came together to discuss how the future of analytics can impact not only ticket prices themselves, but also the fan experience. When asked about how advancements in technology (mobile ticketing, etc) can help teams control the ticket market, Tim Zue (CFO of the Boston Red Sox) made it clear that the secondary market was the future, but these advancements make it safe and easy to offload tickets, not just for brokers, but for season ticket holders as well. What was interesting was that he said the battle against secondary market distributors was over, and that teams need to partner with them to continue to improve the process. This also allows for teams to get access to massive data volumes held by companies like Stubhub for their own analysis on dynamic ticket pricing.

In the past, ticket prices were set by section, and held pretty firm over the course of a season. Yet when you go to buy a ticket on Stubhub, the ticket prices can vary wildly, not just from game to game, but even from seat to seat in the same section. This idea of dynamic pricing has really hit home with the teams, who are moving to apply this type of pricing to the primary market sales as well. Kurt Schwartzkopf, Senior VP of NBA/NHL Arenas for Ticketmaster, noted that dynamic pricing in the primary market not only helps teams to personalize the experience for a new fan who may buy a ticket in the primary market instead of the secondary market, but also helps to limit fraud. If season ticket holders have a safe environment to resell tickets back through a team, the buyer can be sure that this is a safe ticket, and will be guaranteed entrance on game day.

Fan analytics doesn’t just revolve around ticketing, but also can have huge impact on how teams and the media interact with their fans every day. In a session hosted by ESPN, much of the discussion was about how ESPN and the NBA uses analytics to better understand their fan base, and how they interact with their platforms and services. When asked about what they were learning about their fans, Doris Daif, SVP for Customer Data Strategy for the NBA, was clear about the impact on their organization. She talked about how they are using progressive profiling to tailor a particular experience to a particular fan. They are not only looking at their favorite teams and players, but their geography to decide on how to develop their applications and content so that they feel more tailored to that particular fan. And when asked about using analytics to drive content, Paul Sabin, Analytics Specialist at ESPN, made it clear that by analyzing the interests of the fans, not just using ratings, ESPN can make decisions on which games to feature when. This level of analytics can have a huge impact on the future of media, and their relationships with fans.

Business Analytics

At the end of the day, a sports team is a business. How the business runs, though, has many more inputs than the normal every-day business. Success of the team on the field may be the baseline on how well the business is expected to do, but that is far from the only factor. Though, it was interesting to listen to Steve Ballmer, owner of the Los Angeles Clippers, and Nate Silver, founder of the popular stats blog fivethirtyeight.com, talk about the impact a championship has on an organization. When asked about his goals for the Clippers in the short term, Ballmer was clear that his trades of Chris Paul and Blake Griffin were not because he believes that tanking is the only way to find success, but because he wants to rebuild and still compete. At that point, Silver commented that if the Clippers had a championship in recent years, there would be a lot more leeway in how the moves were looked at, both by the media and by their own fans. He used the example of the Celtics, after winning a championship in 2008, had a lot less scrutiny when they traded away the keystones of their championship team, and completely rebuilt within 5 years from 2012–2017.

When it comes to the other parts of the business, sports teams function in very similar ways as other types of businesses. During a session called “Business Analytics 2.0: Redefining the Sports Business”, Brian Lafemina, SVP of Club Business Development for the NFL, talked about how every area of a sports business has untapped capabilities - bringing in large amounts of data which can be used to better the business. He also commented that, just like other businesses, the NFL is trying to get a 360 degree view of their customers, which are all the fans. He recognized that this is not something unique to sports at all, and that the sports world was definitely behind the times. When asked about what type of technology investments need to be made within a sports business, Charlie Freeman, COO of the Orlando Magic, was clear that his first goal is finding the right staff that fits into the analytical mindset needed to drive a business forward, before he can make big technology investments. He was adamant that without the right people, businesses can have all the data, but none of the actual analysis.

Looking to the Future

As the sports landscape continues to evolve, the complexity of analysis becomes much larger. When you think about the maturity curve of analytics on which most business are measured, the sports industry has been stuck at the bottom, looking at descriptive analytics to try to make their organization better. As the analytics departments have become more mature, and the sheer mass of data has begun to grow, the sports industry as a whole has begun to move towards a more predictive, if not prescriptive model of analytics. Teams like the Arizona Coyotes, Boston Red Sox, New England Patriots, and Houston Rockets have been at the forefront, pushing the envelope on the types of analytics they can do, not only on the players they put out on the field, but on the fans that fill the stadium, as well as on the business itself. Within the evolving landscape, there is a lot of opportunity for others on the periphery to invest in their relationships with teams. There were many different sports-related start ups with booths set up at the conference, with even more companies participating in the panels. The opportunities for sports analytics is growing, and driving the sports world into a new frontier of technology, analysis, and engagement.

--

--

James Anderson
Slalom Technology

Sales Engineering Leader @ Snowflake. All opinions expressed are my own.