Overreaction in Football Wagers

Area of Science:

• Sports Analytics
• Statistical Modeling
• Behavioral Economics

Background:

• Football scores are frequently used to assess team performance and predict future outcomes.
• However, score-based evaluations may inaccurately represent true team abilities due to inherent variability.
• The phenomenon of regression to the mean suggests extreme performances are likely to move closer to the average over time.

Purpose of the Study:

• To investigate the extent to which football scores exaggerate true team ability differences.
• To examine whether betting markets adequately incorporate the concept of regression to the mean.
• To provide insights into the predictive limitations of performance scores in sports.

Main Methods:

• Analysis of historical football match data to quantify score-based ability estimations.
• Statistical modeling to assess the degree of regression to the mean in team performances.
• Examination of betting market data to infer market efficiency regarding regression to the mean.

Main Results:

• Football scores tend to inflate the perceived differences between the best and worst performing teams.
• Evidence suggests that extreme performance levels are not as persistent as scores might imply.
• Betting market data indicates a potential under-adjustment for regression to the mean by gamblers.

Conclusions:

• Football scores are an unreliable sole indicator of a team's actual, stable ability.
• Regression to the mean is a significant factor influencing future team performance that is not fully priced into betting markets.
• Understanding this statistical bias can improve performance evaluation and betting strategies.