The triumphs and tragedies of fandom: Emotional arcs in NFL tweets

πŸ“… 2026-07-20
πŸ“ˆ Citations: 0
✨ Influential: 0
πŸ“„ PDF
πŸ€– AI Summary
This study investigates how regional context, game outcomes, and team performance shape the affective dynamics of NFL fans on social media. Leveraging millions of game-related tweets from the 2011–2014 seasons, the authors combine sentiment analysis with geospatial statistics to quantitatively characterize, for the first time, the temporal emotional arc of fans before, during, and after games, and to delineate each team’s β€œfan radius.” Findings reveal that fan sentiment is initially positive pre-game, declines after kickoff, and experiences a modest rebound at halftime; post-game sentiment rises significantly for winning teams but remains persistently low for losing teams. Notably, overall fan sentiment exhibits only a weak positive correlation with team win rates. This work elucidates the temporal evolution and outcome-driven divergence of collective emotion in sports contexts.
πŸ“ Abstract
Online fandom communities influence public opinion toward movies, musicians, and sports teams. Using a corpus of game-referencing tweets, we measure variation in sentiment toward National Football League (NFL) teams driven by geography, game outcomes, and team performance for the 2011--2014 NFL seasons. We estimate a fandom radius for each team, identifying regions where engagement exceeds background levels of discussion. We find sentiment for both winning and losing teams is positive immediately prior to games, drops at kickoff, and rebounds slightly during halftime. After halftime however, the trajectories diverge: Sentiment for winning teams increases toward the end of the game, while sentiment for losing teams remains low, though both end up below their start of game levels. Finally, a comparison between sentiment and win percentage reveals a weak positive relationship, suggesting that while team success contributes to fandom happiness, other factors also influence how fans discuss the NFL on social media. Our work contributes to a growing body of computational social science research that quantifies the many aspects of modern fandom.
Problem

Research questions and friction points this paper is trying to address.

fandom
sentiment analysis
social media
NFL
emotional arcs
Innovation

Methods, ideas, or system contributions that make the work stand out.

fandom radius
sentiment dynamics
computational social science
social media analytics
emotional arcs
πŸ”Ž Similar Papers
No similar papers found.
E
Elisabeth Kollrack
Department of Mathematics & Statistics, University of Vermont, Burlington, VT 05405; Computational Story Lab, University of Vermont, Burlington, VT 05405; Vermont Complex Systems Institute, University of Vermont, Burlington, VT 05405
M
Michael V. Arnold
Computational Story Lab, University of Vermont, Burlington, VT 05405; Vermont Complex Systems Institute, University of Vermont, Burlington, VT 05405; Vermont Advanced Computing Center, University of Vermont, Burlington, VT 05405
Peter Sheridan Dodds
Peter Sheridan Dodds
Professor/Director, Computational Story Lab, Vermont Complex Systems Institute, UVM
LanguageMeaningStoriesSociotechnical PhenomenaComplex Systems
C
Christopher M. Danforth
Department of Mathematics & Statistics, University of Vermont, Burlington, VT 05405; Computational Story Lab, University of Vermont, Burlington, VT 05405; Vermont Complex Systems Institute, University of Vermont, Burlington, VT 05405; Vermont Advanced Computing Center, University of Vermont, Burlington, VT 05405