- On March 1, 2019
Fusion Sport are excited to once again sponsor and contribute to the Sloan Sports Analytics Conference in Boston, rubbing shoulders with some of the world’s best in sports analytics, data science and technology.
With the agenda released and set for the two day conference, our Sports Scientists Tom Crameri and Marcus Colby have had a good hard look and given us their picks for the weekend. Check out the presentations, research papers and workshops they recommend for the sports science enthusiasts.
Friday 1st March
Whistle While You Work: Measuring Intangibles of NBA Officials
8.00am CA West (Room 302) | Monty McHutchen, NBA
NBA referees make more than 4,000 calling decisions each game. But the accuracy of those decisions is only part of what it takes to be a great referee. Come for a candid conversation with Monty McCutchen, the NBA’s Head of Referee Development & Training, to learn about the other performance factors the league considers – from courage to communication – when developing and evaluating referees, as well as the league’s effort to turn those standards of performance into data.
Performance Optimization Strategies that helped Win a Championship
10.15am CA East (Room 312) | Glenn Stewart, High Performance Manager West Coast Eagles
Do a few things brilliantly well rather than try to cover everything. This is the ‘Hedgehog Principle’ as explored in the book Good to Great, which calls on the power of simplicity to elevate the chances of success. In this session, Glenn will take the audience through this Hedgehog mindset for sport and the high performance strategies that led to his team the West Coast Eagles winning the 2018 Australian Football League Premiership. He will explore the four or five things the high-performance unit chose to do exceptionally well, and how they distinguished the great strategies from the good using data aggregated in Smartabase to critically analyze what related to player and team success.
Using Deep Learning to Understand Patterns of Player Movement in the NBA
12.15pm Research Paper Room | Akhil Nistala, YouTube & John Guttag, MIT
In 2011, SportVU fundamentally changed the way that basketball can be analyzed. STATS SportVU utilized a six-camera system installed in basketball arenas to track the real-time positions of players, at 25 frames per second. In this paper, we demonstrate how we can apply deep learning techniques to this data to produce a queryable database of basketball possessions.
Decomposing the Immeasurable Sport: A deep learning expected possession value framework for soccer
2.00pm Research Paper Room | Javier Fernandez, FC Barcelona, Luke Bornn, Sacramento Kings & Dan Cervone, LA Dodgers
What is the right way to think about analytics in soccer? Is the sport about measured events such as passes and goals, possession percentages and traveled distance, or even more abstract notions such as mistakes (to quote Cruyff, “Soccer is a game of mistakes, whoever makes the fewer wins”)? Analytical work to date has focused primarily on these more isolated aspects of the sport, while coaches tend to focus on the tactical interplay of all 22 players on the pitch. Soccer analytics is lacking from a comprehensive approach that can start to address performance-related questions that are closer to the language of the game.
Training Schedule Confounds the Relationship between ACWR & Injury
4.30pm Research Paper Room | Luke Bornn, Sacramento Kings, Patrick Ward, Seattle Seahawks & Darcy Norman, US Soccer
In the past decade, significant efforts have been made to understand injury risk in sport using subjective (i.e. rating of perceived exertion) and objective (i.e. inertial sensor outputs) playermonitoring strategies (Halson, 2014; Bourdon et al., 2017). One metric in particular that has received significant attention is the acute:chronic workload ratio (ACWR). In the past 5 years, numerous academic papers across multiple sports have concluded that ACWR is predictive of injury risk, and as a result the ACWR has become standard practice in professional sports to manage player workloads (Malone et al., 2017a; Malone et al., 2017b; Colby et al., 2017).
In this paper, we use schedule and training data from two professional teams in Serie A (Italian soccer) and the NFL (American football) to show that training schedule confounds the ACWR-injury relationship, leading to overestimates of the magnitude of the relationship and calling into question earlier studies which have ignored this confounder
Unlocking Potential: The Next Generation of Tracking Data
10.15am Pat Summit Room Ballroom | PANEL
Player tracking technologies in the past decade have provided information that was once unattainable—the distance and speed a player runs in a game, effectiveness against certain defenders, the game situations which are most likely to produce a fast break, and so much more. As the technologies behind player tracking improve, much more information can now be captured: the skeleton of every player in every frame; tracking data for every college or high-school game; or data from historical games (e.g. comparing the performance of Jordan, Gretzky or Maradonna against the current greats). What does this evolving category of information mean for sports? How can teams, leagues, media and fans benefit from the next generation and scale of data being captured? This panel brings together thought leaders across sports and the tracking-data space to discuss these questions and explore the implications of this rapidly expanding industry.
Using R for Predictive Modelling
10.15am Workshop Room | Asmae Toumi, Hockey-Graphs
This introductory session will show you how R can be used to build predictive models. Attendees will be able to participate in exercises using various R packages to predict NHL player points using publicly available datasets. No prior knowledge of R is required.
Making the Modern Athlete: A Conversation with David Epstein and Malcolm Gladwell
2.00pm Bill James Room Ballroom AB
Authors Malcolm Gladwell and David Epstein meet again, five years since their initial 10,000 Hours vs. The Sports Gene talk, to resurface the important and long-standing question of whether nature or nurture determines one’s athletic success. In this one-on-one discussion, Gladwell and Epstein sit down to evaluate what has changed, including Epstein’s latest argument that contrary to popular belief, generalists are the ones who actually succeed in a specialized world
DIY Shot Charts
3.15pm Workshop Room | Momin Ghaffar
Explore techniques in data scraping and data visualization. You will be empowered to own a data pipelining process and build your own basketball shot charts.
Visit Fusion Sport
Over the course of the two-day event Fusion Sport will have an interactive area where you can come and chat to our Human Performance experts. We’ll also be educating delegates with our Competitive Advantage Session as indicated above, presented by Glenn Stewart on his 2018 championship winning AFL team, a session not to be missed.
If you are unable to attend the conference this year sign up here and we will share the sessions with you after the event.