This thesis aims to demonstrate that momentum in professional NFL games is real, measurable, and leads to short-term success. Using play-by-play data from over ten NFL seasons, a custom momentum score system was developed by combining engineered features and changes in win probability for key events. Momentum is defined conceptually and operationally to ensure consistent measurement across games. Different predictive models were then trained to identify momentum shifts, and an event-based validation strategy was implemented to verify the momentum shifts led to successful events. The findings indicate that the momentum score system captures shifts in game control and differs from traditional win probability metrics. Unlike win probability, momentum does not guarantee victory. It can also be gained by either team regardless of the game's expected outcome, builds progressively, and can be stalled rather than entirely lost. Furthermore, with over a 60 percent validation rate of short-term success following a momentum shift, these findings provide evidence that short-term success driven by momentum is not random. This work offers a new quantitative framework for evaluating momentum in football, with implications for understanding game dynamics and team strategies.
","abstract_html":"<p>This thesis aims to demonstrate that momentum in professional NFL games is real, measurable, and leads to short-term success. Using play-by-play data from over ten NFL seasons, a custom momentum score system was developed by combining engineered features and changes in win probability for key events. Momentum is defined conceptually and operationally to ensure consistent measurement across games. Different predictive models were then trained to identify momentum shifts, and an event-based validation strategy was implemented to verify the momentum shifts led to successful events. The findings indicate that the momentum score system captures shifts in game control and differs from traditional win probability metrics. Unlike win probability, momentum does not guarantee victory. It can also be gained by either team regardless of the game's expected outcome, builds progressively, and can be stalled rather than entirely lost. Furthermore, with over a 60 percent validation rate of short-term success following a momentum shift, these findings provide evidence that short-term success driven by momentum is not random. This work offers a new quantitative framework for evaluating momentum in football, with implications for understanding game dynamics and team strategies.</p>","abstract_has_math":false,"creators":["Zahller, Sean M"],"institution":null,"degree_name":"Master of Science (MS) in Computer Science","degree_level":"Thesis","degree_discipline":"Computer Science and Electrical Engineering","degree_department":null,"school":null,"contributors":["Dan Li"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-01-01T08:00:00Z","date_published":"2025-01-01T08:00:00Z","updated_at":"2026-07-24T02:13:16Z","subjects":["Computer Science","Data Analysis","Machine Learning","Big Data","Sports Analytics","Sports Studies"],"languages":[],"rights":["Access is available to all users"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://dc.ewu.edu/theses/989","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Dan Li"]},{"key":"dc:creator","label":"Author","values":["Zahller, Sean M"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science and Electrical Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science (MS) in Computer Science"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Computer Science","Data Analysis","Machine Learning","Big Data","Sports Analytics","Sports Studies"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["Access is available to all users"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://dc.ewu.edu/theses/989"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This thesis aims to demonstrate that momentum in professional NFL games is real, measurable, and leads to short-term success. Using play-by-play data from over ten NFL seasons, a custom momentum score system was developed by combining engineered features and changes in win probability for key events. Momentum is defined conceptually and operationally to ensure consistent measurement across games. Different predictive models were then trained to identify momentum shifts, and an event-based validation strategy was implemented to verify the momentum shifts led to successful events. The findings indicate that the momentum score system captures shifts in game control and differs from traditional win probability metrics. Unlike win probability, momentum does not guarantee victory. It can also be gained by either team regardless of the game's expected outcome, builds progressively, and can be stalled rather than entirely lost. Furthermore, with over a 60 percent validation rate of short-term success following a momentum shift, these findings provide evidence that short-term success driven by momentum is not random. This work offers a new quantitative framework for evaluating momentum in football, with implications for understanding game dynamics and team strategies.
"]},{"key":"dc:title","label":"Title","values":["Identifying and Validating Momentum Shifts in the NFL A Data-Driven Approach to Predicting Short-Term Team Success"]}]}],"canonical_facts":{"dc:contributor":["Dan Li"],"dc:creator":["Zahller, Sean M"],"dc:description.abstract":["This thesis aims to demonstrate that momentum in professional NFL games is real, measurable, and leads to short-term success. Using play-by-play data from over ten NFL seasons, a custom momentum score system was developed by combining engineered features and changes in win probability for key events. Momentum is defined conceptually and operationally to ensure consistent measurement across games. Different predictive models were then trained to identify momentum shifts, and an event-based validation strategy was implemented to verify the momentum shifts led to successful events. The findings indicate that the momentum score system captures shifts in game control and differs from traditional win probability metrics. Unlike win probability, momentum does not guarantee victory. It can also be gained by either team regardless of the game's expected outcome, builds progressively, and can be stalled rather than entirely lost. Furthermore, with over a 60 percent validation rate of short-term success following a momentum shift, these findings provide evidence that short-term success driven by momentum is not random. This work offers a new quantitative framework for evaluating momentum in football, with implications for understanding game dynamics and team strategies.
"],"dc:identifier":["https://dc.ewu.edu/theses/989"],"dc:rights":["Access is available to all users"],"dc:subject":["Computer Science","Data Analysis","Machine Learning","Big Data","Sports Analytics","Sports Studies"],"dc:title":["Identifying and Validating Momentum Shifts in the NFL A Data-Driven Approach to Predicting Short-Term Team Success"],"thesis:degree_discipline":["Computer Science and Electrical Engineering"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["Master of Science (MS) in Computer Science"]},"updated_at":"2026-07-24T02:13:16Z"}