Post-Graduate Stories

Ken Peng Will Develop and Implement “AI-Powered Statistical Analysis for Elite Team Sports”

As a 2026 CANSSI Distinguished Postdoctoral Fellow, Kangyi (Ken) Peng will take part in a comprehensive program that involves teaching, interdisciplinary or applied collaboration, professional development, and a research project focusing on the development of a framework that integrates advanced statistical methodology and cutting-edge artificial intelligence (AI) to address fundamental challenges in elite team sports. Ken will work under the supervision of Professors Tim Swartz (Simon Fraser University) and Tianyu Guan (York University).

Program: CANSSI Distinguished Postdoctoral Fellowships
Region: National
Date: 2026–2028

Project Focus Areas

Ken’s research program will develop a novel framework that integrates advanced statistical methodology and cutting-edge artificial intelligence to address fundamental challenges in elite team sports. It pioneers the integration of graph neural networks (GNNs) and Markov decision processes (MDPs) to create prescriptive models for optimal strategy in dynamic and network-based environments. Furthermore, it breaks new ground by fusing large language models (LLMs) with causal inference for coaching impact analysis and with Bayesian hierarchical models for lineup and rotation evaluation.

Ken will lead the development and implementation of these models, transforming multimodal data into sports insights. He will conduct research at the intersection of statistics, AI, and sports analytics. The work provides significant methodological contributions in statistical learning and direct applications for elite team sports programs.

In addition to engaging in research, Ken will teach undergraduate statistics courses and mentor/co-supervise undergraduate and graduate students, while participating in professional development opportunities at Simon Fraser University and York University.

Getting to Know Ken

Ken Pang received his PhD in Statistics from Simon Fraser University in 2025. His research was supervised by Professors X. Joan Hu and Tim Swartz and focused on statistical modelling for dynamic and latent-process systems, with applications in environmental public-health surveillance and sports analytics. He is broadly interested in Bayesian methods, stochastic-process modelling, and biostatistical methodology.

He is currently a postdoctoral fellow working with Professor Robert Delatolla’s group at the University of Ottawa to develop spatio-temporal models for wastewater-based surveillance to support public-health decision making. Beyond academia, he is the co-founder and chief data scientist of AccMov, a sports technology start-up based in Ottawa.

Ken states that his “long-term aspiration is to become an independent statistical researcher whose work is driven by real-world questions while also contributing to statistical methodology.” He also identifies teaching as “an integral part of [his] career and hope[s] to train the next generation of statisticians to combine rigorous reasoning with modern tools and applied perspectives.”

Ken sees the CANSSI Distinguished Postdoctoral Fellowship as an opportunity “to consolidate my existing research into a coherent program, while allowing me to explore new application domains and methodological directions.”

Through CANSSI’s collaborative network, I [will] expand my connections, gain broader exposure to research problems, and strengthen my teaching experience. The fellowship [will] also enable me to develop statistical software, helping translate methods into tools that can be used by the wider research community.