Hackathon

In professional basketball, it is crucial for the coaching staff of a team to analyze an opposing team and develop an effective strategy. Understanding player shooting profiles is an essential part of this analysis: knowing where certain opposing players like to shoot from can help coaches neutralize offensive gameplans from their opponents, while understanding where their players are most comfortable can lead them to developing more effective offensive strategies. We present a tool that can visualize player performance profiles in a timely manner while taking into account factors such as play-style and game dynamics, generating interpretable heatmaps that allow us to identify and analyze how these non-spatial factors affect the performance profiles. Our methods provide an effective and efficient tool that can provide insight into how certain players and teams play, without requiring the time-consuming process of reviewing hours of film, and could potentially be applied to other sports with adaptations.

As an amateur goalkeeper playing grassroots soccer, who better to learn from than top professional goalkeepers? In this paper, we harness computer vision and machine learning models to appraise the save technique of professionals in a way those at lower levels can learn from. We train an unsupervised machine learning model using 3D body pose data extracted from broadcast footage to learn professional goalkeeper technique. Then, an “expected saves” model is developed, from which we can identify the optimal goalkeeper technique in different match contexts.

This paper applies traditional sports analytics metrics with novel machine learning models in a brand new competitive Esport. By leveraging in-game positional data, we are able to evaluate the difficulty of a particular gun fight and assign a win probability to both sides. We use these predictions to identify players who are performing above or below expected, and identify strengths and weaknesses for NRG’s player development. We are hopeful for more analytics in Esports from current working professionals and the younger generation.

This paper draws upon recent advances in Natural Language Processing (NLP) and Computer Vision (CV) to learn to describe the way in which players impact the game in the MLB. In particular, this work views the game as a sequence of events - instead of a set of summary statistics describing said events - and trains machine learning models to describe the impact that a given sequence of events has on the game. The models describe a sequence of events for a single player over a relatively small time period; so we refer to the model output as player form embeddings - descriptions of how they have impacted the game in the short term. We demonstrate how these embeddings can be used to describe players over the short- and long-term, and contain signals useful for predicting the outcome of games.