Sports Analytics Weekly by kubeia.io - 12/2024

The wagering situation involving Shohei Ohtani’s former interpreter shows that people adjacent to the players and coaches can also be a risk.

The Cowboys recently exercised the 5th year option on All Pro EDGE defender Micah Parsons. Nothing really of note there, Parsons has been one of the top defenders in the league for a while now. The interesting part arises when looking at how the Cowboys went about it. The Cowboys designated Parsons as a Defensive End instead of a linebacker, saving just over $2.5 million. In this article about Edge defenders, we luckily don’t have to nitpick about outdated position names and can just analyze four of the top prospects for the draft: Dallas Turner, Jared Verse, Laiatu Latu, and Chop Robinson.

For the second time this Premier League season, a points deduction for breaching its profit and sustainability rules (PSR) has dragged a club down the table and into the relegation zone.

Jeremy Steele is the Managing Director of sports analytics consultancy, Analytics FC. He is quite unique in the fact that he has experience as a coach, head coach, scout, sporting director as well as a his current business leadership role.

Analytics FC work with clubs all around the world but also work with players too, with one notable project being the work they did with Kevin De Bruyne when he negotiated his current long term contract with Man City.

The system can predict the outcome of corner kicks and provide realistic and accurate tactical suggestions in football matches.

Abstract:Time series forecasting plays a crucial role in decision-making across various domains, but it presents significant challenges. Recent studies have explored image-driven approaches using computer vision models to address these challenges, often employing lineplots as the visual representation of time series data. In this paper, we propose a novel approach that uses time-frequency spectrograms as the visual representation of time series data. We introduce the use of a vision transformer for multimodal learning, showcasing the advantages of our approach across diverse datasets from different domains. To evaluate its effectiveness, we compare our method against statistical baselines (EMA and ARIMA), a state-of-the-art deep learning-based approach (DeepAR), other visual representations of time series data (lineplot images), and an ablation study on using only the time series as input. Our experiments demonstrate the benefits of utilizing spectrograms as a visual representation for time series data, along with the advantages of employing a vision transformer for simultaneous learning in both the time and frequency domains.

Abstract:Black-box machine learning models are now routinely used in high-risk settings, like medical diagnostics, which demand uncertainty quantification to avoid consequential model failures. Conformal prediction is a user-friendly paradigm for creating statistically rigorous uncertainty sets/intervals for the predictions of such models. Critically, the sets are valid in a distribution-free sense: they possess explicit, non-asymptotic guarantees even without distributional assumptions or model assumptions. One can use conformal prediction with any pre-trained model, such as a neural network, to produce sets that are guaranteed to contain the ground truth with a user-specified probability, such as 90%. It is easy-to-understand, easy-to-use, and general, applying naturally to problems arising in the fields of computer vision, natural language processing, deep reinforcement learning, and so on. This hands-on introduction is aimed to provide the reader a working understanding of conformal prediction and related distribution-free uncertainty quantification techniques with one self-contained document. We lead the reader through practical theory for and examples of conformal prediction and describe its extensions to complex machine learning tasks involving structured outputs, distribution shift, time-series, outliers, models that abstain, and more. Throughout, there are many explanatory illustrations, examples, and code samples in Python. With each code sample comes a Jupyter notebook implementing the method on a real-data example; the notebooks can be accessed and easily run using our codebase.