Skip to content
Sean Clement
Sean Clement

Sean Clement

Director of Analytics and Strategy
New York Jets
Bio

Sean Clement is the Director of Football Analytics and Strategy at the New York Jets. Previously, he was the Vice President of Football Operations at SumerSports, and has worked as an analytics consultant for the Baltimore Ravens, Miami Dolphins, and Cincinnati Reds where he built player evaluation models using tracking data, scouting grades, and outside data sources to improve player acquisition and development. Sean is a retired veteran of the U.S. Army where he served with distinction overseas in both Europe and Afghanistan as UH-60 Blackhawk pilot and later as an Operations Research Systems Analyst (ORSA). He was recognized as the analyst of the year by both the Military Operations Research Society and the Army Operations Research Society in 2020 before medically retiring in 2023.

Sean graduated from the United States Military Academy at West Point with a bachelor’s degree in operations research in 2009 and Stanford University with a master’s in computational mathematics and engineering in 2018.

Sean, his wife Kaitlyn, and their daughters Eleanor and Madeline currently live in New Jersey.


Building a High-Trust Culture that Ensembles Data and Judgment in Football and the Army

Organizations often treat “becoming data-driven” as a technology problem: acquire more data, hire analysts, deploy dashboards, and trust the model. In talent management, that approach often falls short; not because the data lack value, but because the best decisions fuse evidence with experienced judgment and local context. In this talk, I draw on experiences from football analytics and the U.S. Army to argue that successful human capital optimization depends on a cultural framework that fuses quantitative analytics with qualitative expertise at every decision point.

I propose a practical reframing: treat human expertise as models themselves. They are learned systems with priors, features, and failure modes. When we acknowledge that the expertise and mental frameworks of scouts, coaches, commanders, and operators are “models,” we can evaluate, calibrate, and ensemble their judgments with statistical and machine-learning outputs rather than positioning them as opponents. This lens also clarifies a common pitfall: a hyper-focus on bias reduction can increase variance, just as it can when over-regularizing a model or pruning an ensemble. The goal is not bias-free decision making; it is reliable decision making under uncertainty.

Finally, I outline how AI can be used to condense human intelligence rather than replace it; capturing expert context, surfacing assumptions, and accelerating learning loops. The talk closes with a culture playbook: build trust, ask better questions, invite principled pushback, and normalize conflict as good-faith error checking. The result is a decision environment where analytics and expertise co-lead and where talent choices improve because the culture makes the models better.

Professional