Investigating Video Assistant Referees (VARs): Does contextual information influence VAR decision-accuracy and decision-severity in football.

Football is the most popular sport in the world, with large financial investment at the forefront of the sport which is evident from the transfer prices of players at the top level of the game. Errors are therefore extremely costly to many stakeholders within the sport. In comparison to other sports, football is behind in terms of technological advancements, with Rugby having a video referee (TMO) since 2001, who has the aim of assisting the referee when incorrect decisions occur. In football, VAR has been recently implemented to reduce errors made by on-field officials but given that this technology is still operated by referees, external influences such as contextual information could allow for errors to still occur. An overview of the decision-making and officiating literature (Chapter 1) crucially highlights that many factors have been shown to influence the on-field referees decision-making but limited research into VAR has occurred due to its infancy in football. A scoping review was conducted (Chapter 2) to synthesis current knowledge surrounding VAR which will guide this new area of research forward. An experimental study (Chapter 3) was also conducted, bringing together findings from chapter 1 and 2, which investigated the effect that contextual information has on the decision-accuracy and severity of referees during a process mimicking VAR, utilising an online video paradigm to which allowed the referees to view multiple replay angles. Findings from chapter 2 indicated that VAR was in its infancy, with limited studies focusing on specific match variables interest. The findings suggest that VAR does exert influence over few specific match variables, which include increases in match duration and decreases in fouls, yellow cards, red cards and offsides. Some of these variables have been suggested to mediate the influence of home advantage. Additionally, no primary research had been conducted to date, with many gaps’ worth exploring to aid stakeholders within football. In chapter 3 no statistically significant difference between contextual groups were found for the experimental study, with low decision-accuracy across all groups, which was attributed to the level of referee recruited for the study. No statistically significant difference was found for decision severity either, which may have been a result of the cognitive load of the contextual information provided to the referees. This thesis expands on the literature relating to VAR; however, further research is needed to identify the potential influence that contextual information can exert over VAR referees specifically.