The MIT Sloan Sports Analytics Conference is more of a sports business conference these days, but the MIT association conveys a conference that also values rigorous and high-quality research.  The Research Paper competition keeps the SSAC connected to this heritage as it evolves.  This year the competition returns with tracks for publications on Baseball, Basketball, Other Sports, and the Business of Sports.

The competition is in two phases: the first phase whittles down the field on the basis of research abstracts submitted, and the survivors are invited to submit full papers which are then reviewed by the SSAC jury.  The eight finalists present their papers in the Research Paper session on Friday and Saturday.  The winning submission receives a substantial cash prize and a guest column on ESPN’s FiveThirtyEight website.  You can view the finalists and their papers on the SSAC website.

Basketball and baseball are always represented among the finalists (there were two finalists from each sport), but there are also tracks for other sports and the business of sports.  The remaining finalists represent fantasy sports, eSports (for the first time, I believe), American football, and, for the fifth year in a row, soccer.

It is extremely interesting to observe how closely the Research Paper competition resembles my experience with technical conferences in STEM fields. Many of the finalists are repeat SSAC finalists, some presented posters in previous years, and others appeared at other sports analytics conferences. Almost all of the finalists have some kind of institutional support and are either employed by or collaborate with sports data companies.  It may still be possible to present a poster at SSAC as an independent researcher, but those days are long past for the research paper competition.

The finalists:

• Extending the word2vec algorithm to model similar and orthogonal baseball talent [Very interested in this one]
• Relating sensorimotor abilities to on-field performance using Bayesian modeling
• Extending defensive player ghosting to basketball play planning [An extension of last year’s presentation by the Patrick Lucey’s group]
• Evaluating basketball shot mechanics and performance using a unique high-resolution data set [An extension from last year’s Poster Competition]
• Relating training volume and intensity to injury rate in American football using Bayesian analysis
• Space creation metric development in soccer [Very interested, but this work looks familiar…]
• An open-source eSports data and analytics platform [Always interested in data platform projects]
• Winning strategies for fantasy sports [Applying operations research to fantasy sports decision-making]

This year’s papers reflect either the continuance or emergence of a few trends in sports analytics research:

• Natural language processing (NLP) techniques applied to sports data modeling
• Deep learning applied to player tracking data
• The increased application of Bayesian modeling
• Analytics projects released to open source

If you’re attending the conference, check out the Research Paper presentations. It’s been a showcase for emerging sports analytics research and a springboard for future careers in the field.

I’ll preview the Research Posters in a separate post.