Abstract
NFL teams may have an advantage over their opposition when players are more familiar with game-day temperatures; however, relationships between game-day air temperature and team performance had not been examined using official at-stadia air temperatures. Game data from 2017 to 25 were collected from NFL "Gamebooks." Data were extracted for games contested by a team based in a cool climate >39° N (North teams, n = 17) against a team based in a warm climate <39° N (South teams, n = 15). Data extracted included game-day temperature (°C), game location (North, South), game outcome (win, no-win [relative to North team]), and point differential (points [relative to North team]). Mixed effects logistic regressions examined whether game-day temperature associated with game outcome, while mixed effects models examined whether temperature associated with point differential. Models were run for all games combined, and for games contested in North and South locations, respectively. North team win probability declined as game-day temperature increased (OR = 0.974, 95% CI [0.951, 0.997], p = .026), and North team point differential was negatively associated with temperature (β = -0.175, 95% CI [-0.286, -0.063], p = .002). Models run according to game location revealed that North team point differential was associated with game-day temperature for games played South (β = -0.152, 95% CI [-0.242, -0.063], p = .001). In NFL games contested by North (>39° N) versus South (<39° N) teams, North teams are less likely to win and have worse point differentials, as game-day temperatures increase. Research should explore whether behavioral or psychological adaptations to familiar game-day temperatures explain these findings.