From professor to founder, statistically grounded expertise in
predictive modeling, machine learning, and AI.
I’m Dr. Ian Barnett. Throughout my career, from academia to consulting and entrepreneurship, I have devoted myself to developing and applying cutting-edge methods in predictive modeling, longitudinal data analysis, high-dimensional statistics, network science, machine learning, and AI. My research spans applications from biostatistics to sports analytics.
The common through-line: when there’s a problem and pertinent data available,
I will develop the solution. Try me!
I founded Seam, a sports analytics company, to provide the front offices of professional sports teams guidance for decision making related to player acquisition, roster management, and in-game strategy based on cutting-edge predictive modeling. The flagship product is the Generative AI MLB Engine (GAME), based on 27 in-house models (from standard statistical models to neural nets/transformers) for each micro-event that can occur in a game. I have also consulted for multiple sports analytics companies where I directed statistical design of projects ranging from injury prevention to player evaluation.
My research focuses on developing statistical, ML, and deep learning/AI methods for complex longitudinal, correlated, and high-dimensional data, with applications in biostatistics and sports analytics. I have a hands-on approach to research where even when mentoring research projects I enjoy diving into the data myself. I have 52 peer-reviewed publications, including 20 as first or last author. I was PI of a >$1M R01 NIH grant (“Statistical methods in mHealth to signal interventional needs for mental health patients”), led several internal seed grants at Penn, and contributed to numerous NIH awards as a co-I.
I have a long history of mentoring students from undergraduate to postdoctoral. At the University of Pennsylvania (Department of Biostatistics, Epidemiology & Informatics) I supervised 31 mentees:
In addition, I chaired 6 PhD dissertation committees and served on 9 others.
At the University of Pennsylvania, I taught the PhD-level Longitudinal Data Analysis course in the Department of Biostatistics, Epidemiology, and Informatics for six years, earning consistently strong student evaluations (average 3.7/4.0). I organized and led the Digital Health working group, and frequently presented at the Wharton Sports Analytics and Business Initiative working group as well as at Wharton Moneyball Academy. In addition to numerous guest lectures, I routinely present at national and international conferences spanning statistics, biostatistics, sports analytics, AI and related fields.
