Exploring the socio-technical dynamics of VAR implementation and use

ABSTRACT

VAR (Video Assistant Referee) is one of the most significant and controversial developments in world football. At its core, VAR is a decision-support system that aids referees to make more accurate decisions and eliminate clear and obvious errors. While current research highlights an increased level of accuracy, there have also been highly publicised errors, with fans having a mixed response. In an effort to gain a holistic understanding of VAR, this study undertakes a grounded theory approach to capture the socio-technical dynamics that have dominated the VAR implementation and use. The study focuses on the English Premier League (EPL) and examines over 2 years of news articles on VAR. Using open coding, 14 categories emerge, providing a comprehensive picture of the socio-technical dynamics. These categories are organised into ‘Socio-Technical Context of the VAR Implementation’ and ‘Socio-Technical Impact of the VAR Implementation’. The paper concludes by stating several propositions that can be explored further.

KEYWORDS:

• Word
• socio-technical analysis
• grounded theory
• video assistant referee
• sport decision support

Introduction

VAR, or video assistant referee, incorporates the use of real-time video and replays to assist the referee to make decisions during critical incidents during football games. The on-field referee is assisted by a VAR team who monitor multiple screens and can control video playback and camera angles to revisit in-game incidents (see Figure 1 for a DALLE generated image of a VAR room description). First conceived in 2010 by the Royal Netherland Football Association (KNVB) as part of their Refereeing 2.0 project, it took almost a decade for the technology to be included in the FIFA ‘Laws of the Game’.¹ Argued as one of the most significant developments in world football since 2018, VAR has gradually been used in all major domestic and international competitions, including the men’s and women’s World Cups, the Champions League, the Copa Libertadores, and the Club World Cup (Zglinski, 2022).

Figure 1. DALLE generated image from the description of Stockley Park outlined below.

While the use of VAR is still at an early stage (de Oliveira et al., 2023), it has attracted much attention from researchers and the general public. A key focus for research has been on the effectiveness of VAR (Errekagorri et al., 2020; Han et al., 2020; Meneguite et al., 2022; Spitz et al., 2018), and fan perceptions of the decision support system (Hamsund & Scelles, 2021; Kolbinger et al., 2020; Scanlon et al., 2022; van den Berg & Surujlal, 2020; Winand et al., 2021). From an effectiveness perspective, research highlights that post-VAR; there has been a reduction in the number of fouls, offsides, and yellow cards (Han et al., 2020; Lago- PeñPeñAs et al., 2019; Meneguite et al., 2022) but an increase in playing time (Han et al., 2020). Interestingly, VAR has only a very minor impact on home team advantage (Han et al., 2020), demonstrating the pre-existing high level of officiating capability (Holder et al., 2022), but officials do tend to penalise situations more severely in slow motion compared to real-time (Spitz et al., 2018). Nonetheless, other research found that the changes are all minor and ‘VAR hardly changes the game’ (Errekagorri et al., 2020, p. 808), with its impact mainly limited to the rules and not the norms of the game (Zglinski, 2022). Finally, VAR is predicted to highlight weaknesses in the game’s rules, which may need to change in the future (Tamir & Bar-Eli, 2021).

From a fan perspective, research has highlighted a complex relationship between VAR and fans (Scanlon et al., 2022). This includes a fragmented perception, differing perspectives, and differing experiences of the influence of VAR (van den Berg & Surujlal, 2020; Winand et al., 2021). For instance, one study found that the majority of fans were happy with VAR, with younger groups demonstrating more positive perceptions than their older counterparts (Hamsund & Scelles, 2021). Conversely, in an assessment of public sentiment through X (formerly Twitter), VAR was found to be associated with predominantly negative sentiments (Kolbinger et al., 2020).

Summarising the numerous studies on VAR research, they have mostly taken a narrow research focus and presented somewhat conflicting results from a sports, behavioural, and decision-making perspective. In response, we aim to develop a more holistic understanding of VAR from an Information Systems perspective. More specifically, the research objective of the study is to identify the socio-technical dynamics that characterise VAR implementation and use. As a result, the study will provide context to the many existing studies on VAR and ultimately explain the complexities and conflicting results of previous research.

To fulfil this objective, this paper reports on our grounded theory approach to analyse 2 years of English Premier League (EPL) news articles that focus on VAR. The source of these sports analysis articles came from a freely accessible international news media organisation (The Guardian). This secondary data source is further discussed in the next section with the open coding technique used as part of our Methodology. Our efforts, presented in the Findings section, produce 14 categories, organised across the ‘Socio-Technical Context’ and ‘Socio-Technical Impact of the VAR Implementation. We argue that these 14 categories present a holistic insight into the implementation and use of a simple decision support technology in a complex socio-technical system/environment. The higher-level categories of ‘Socio-Technical Context of the VAR Implementation’ and ‘Socio-Technical Impact of the VAR Implementation’ position the emerging 14 sub categories into the existing context before the implementation of VAR and the impact of VAR. As a result, this provides greater insight into the source and impact of the socio-technical dynamics evident within the study.

Methodology

Taking a grounded theory approach, this paper aims to capture the socio-technical dynamics occurring within the implementation and use of VAR. We view grounded theory as ‘a practical method for conducting research that focuses on the interpretive process by analysing the actual production of meanings and concepts used by social actors in real settings’ (Suddaby, 2006, p. 633). Furthermore, a grounded theory approach is most appropriate when the research question posed by the researcher aims ‘to make knowledge claims about how individuals interpret reality’ (Suddaby, 2006, p. 634). Therefore, as popularised by Wolfswinkel et al. (2013, p. 2), the inductive nature of grounded theory ‘lets the salient concepts arise from the literature’. Within the Information Systems (IS) discipline, grounded theory is an accepted research approach (c.f. Alhassan et al., 2023; Seidel & Urquhart, 2013) and is often characterised by theory generation from the ‘continuous interplay between the collection and analysis of data’ (Seidel & Urquhart, 2013, p. 237). Coding procedures are central to grounded theory, and the Strauss and Corbin strand of grounded theory is ‘arguably the most influential’ in the IS discipline (Seidel & Urquhart, 2013, p. 238), where the coding process is broken into ‘three stages of open, axial, and selective coding’ (ibid., p.239).

Data collection

As VAR has been implemented in numerous leagues and competitions across the globe, however, following a purposeful sampling strategy which is suited to constructing multi‐perspectival, emancipatory, participatory and deconstructive interpretations (Suri, 2011), it was decided to focus our data collection on a specific league. The English Premier League is noted as one of the most prestigious leagues in the world and one in which VAR is receiving significant criticism, thus providing a rich source of publicly accessible data. To provide a consistent, relevant, rich and open source of data, we decided to collect media articles from The Guardian (https://www.theguardian.com) news outlet. The Guardian is a freely accessible online paper that also has a specific section dedicated to VAR (https://www.theguardian.com/football/video-assistant-referees-vars). Covering 24 months (Dec 2021 – Dec 2023), 50 articles were identified within the section and validated as being articles that focus on VAR. To cross-check the list of articles, a separate search was conducted through the newspaper’s search interface, which added two further articles to the list. The use and acceptance of newspapers as a primary source of data on research studies is increasing across many academic disciplines (Meyer, 2018). Providing multiple perspectives, the articles cover all aspects of VAR usage and its implementation through editorials, analysis, match reports, interviews with managers, players, and referees, and voicing public opinion.

Data analysis

With the aim of gaining a wide and holistic understanding of the complexities of the VAR implementation, we followed an open coding approach (as the first stage in our grounded theory approach). In open coding, categories are identified and labelled through an examination of the data. However, these categories emerge from a grouping of initial codes with similar meaning (Alhassan et al., 2023). Therefore, this involves breaking down the data into smaller units (e.g. sentences and paragraphs) and assigning ‘initial codes’ to these ‘chunks of data’ (Seidel & Urquhart, 2013, p. 239). Within grounded theory, this process is typically inductive and fuels the emergence of observable patterns and themes from the set of initial codes generated. Therefore, the goal of open coding is to be comprehensive and open-minded in capturing the essence of the text, and this allows us to see what might not always be obvious in the data. In this study, every relevant (VAR-focused) article sentence/section was analysed and tagged with an initial code. This coding was completed by the lead author, and the coded outputs were reviewed by the other research team members through ‘show and tell’ conversations throughout the open coding process. In total over 436 article sections were analysed, generating 101 different codes and 1,481 coded instances within the collection of texts (50 Guardian articles). Initially, as codes emerged, they were also categorised under the headings of people, process, and technology, which are the fundamental socio-technical components of an Information System (c.f. Keen, 1993). Toward the later stages of the open coding process, these three socio-technical components informed the four categories placed within the ‘Socio-Technical Context’. Furthermore, the 10 categories within ‘Socio-Technical Impact’ reflect more on the socio-technical dynamics of these three components (e.g. how people, process, and technology all come together). In the next section, we present the outputs of our open coding efforts (14 categories) organised into ‘Socio-Technical Context’ (4 categories) and ‘Socio-Technical Impact’ (10 categories).

Findings: emergent categories

Socio-technical context of VAR implementation

Low automation

Compared to the level of complexity that decision support technologies can attain, VAR is a simple technology that involves video replay capabilities that are manually monitored by Video Assistant Referees in an offsite location at Stockley Park. Described in detail by one journalist ‘the VAR room is large, and with its windowless walls, dimmed lighting and ergonomic chairs, it looks a bit like a presidential war room. There are 10 individual VAR stations, each with three seats, so multiple matches can be VAR’d at the same time. During the match, the VAR sits in the middle, with an assistant on the left and a replay operator on the right. A screen in front of each seat plays a live feed of the match, and a second screen shows it from four different camera angles. If the VAR identifies what they believe might be a clear and obvious error, they press a red button, which opens up a channel with the referee, to instruct them to pause to allow for a check’ (Article 37). Except for automatically assessing if the ball has crossed the goal line, all other decisions require human intervention. Indeed, ‘PGMOL has ruled out using offside-only VARs’ (Article 47), which would have fully automated all offside calls, without needing VAR reviews.

Stakeholder complexity

The analysis identified five different groups of stakeholders, each with multiple subgroups and instances. As a result, this provides a rich environment for a high-level of socio-technical dynamics in the context of VAR.

Officials

These are ultimately the users and operators of the technology and see themselves as the ‘21st team in the division’ (Article 37). Along with the introduction of VAR, a new role titled the Video Assistant Referee was also created, which is supported by two or more Assistant VARs that are observing footage and inputting into decision support. Highlighting the socio-technical aspect of the decision support system, when VAR is mentioned, it tends to reference the entire system rather than the actual Video Assistant Referee. The referee is the key stakeholder that VAR is there to support, along with the sideline officials who also play a role in officiating the game in real-time. These stakeholders have ultimate responsibility for the decisions being made and are often removed for selection if an error has occurred as evidenced in the following ‘the first weekend of the season was marred by match officials failing to award Wolves a stoppage-time penalty at Manchester United, leading to Hooper, also the referee that day, Michael Salisbury, the VAR, and Richard West, the AVAR, being dropped from duty for the following round of matches’ (Article 24).

Fans

Fans are the consumers of the end product that is professional football. Within the dataset three types of fans are mentioned in connection with VAR. There are fans that are in the stadium, watching the game live who are ‘often forgotten in all of this’ VAR analysis and discussion. There are fans watching live but through mediated channels (e.g. TV, radio, online updates) who ‘are fed a stream of instant punditry and replays’ (Article 1) about VAR decision support. There are also general fans the ‘the football-watching public’ (Article 110) who may not always watch live matches but get match reports or highlights to keep up to date with their chosen club and football team. They usually get a retrospective analysis of the game and of the VAR decision support without witnessing the live event.

Clubs/managers/players

These are stakeholders directly affected by the decisions made by officials during the game and are regularly mentioned in connection with VAR. The proximity of this group to the officiating is underlined by the highly competitive nature of the league, which means that official decisions can not only determine the outcome of a game but also the outcome of a full season. Of this group, managers are much more vocal on the topic of VAR. In our data-set, players are predominantly mentioned as part of the decision but are not directly quoted on VAR that often. In contrast, managers are consistently quoted and articulate a range of emotions, from sympathy for the referee to confusion, anger, and criticism for the decisions being made, which have usually gone against their team. Clubs are mentioned when formal appeals or requests are being made to governing bodies.

Governing bodies

Within professional football, there are a range of governing and representative bodies that look after different aspects of the game. These include IFAB (the International Football Association Board), who are independent guardians of the football rules. They essentially created the VAR protocol, which paved the way for VAR to be implemented across all major professional leagues and competitions. Within the Premier League competition, there is the PGMOL (Professional Game Match Officials Limited), which is a non-profit service company that manages officiating and is responsible for the day-to-day implementation of VAR (e.g. staffing, protocol interpretation, benchmarking). PGMOL is the most mentioned body in our data set, with IFAB coming a distant second. Other bodies that are mentioned include the Premier League (body) which is responsible for running the competition itself, the FA (Football Association) which has veto power over any changes to the rules, UEFA (Union of European Football Associations) the governing body of football in Europe, FIFA (Fédération internationale de football association) the international governing body of football, and the LMA (League Managers Association) the representative body of the managers.

Pundits/journalists

The final group that is identified in the dataset are football journalists and pundits. Pundits are primarily past players but have been recently joined by past referees to ‘to put all the weekend’s controversies under a microscope’ (Article 110). The role of the pundit is to offer their opinion and in the case of former footballers they have been for the most part against VAR and critical of mistakes being made. For instance, ‘’My 12-year-old would be better than some of the decisions I’ve seen this season’, said former player and pundit Danny Murphy’ (Article 37). Given the dataset was taken from The Guardian newspaper, opinion and insight from football analysts/journalists were also captured during the coding process. The big difference between this group and pundits (as defined in this research) is that journalists are the primary generators of the text being analysed and are not ex-players.

Decision difficulty

‘Football refereeing is one of the toughest jobs in sport, and over the years, it takes its toll. “I can’t think of any other job where there’s this much scrutiny over decision-making. Maybe a surgeon?” said Marriner [Referee] recently’ (Article 37). This difficulty is also acknowledged by other stakeholders, Ange Postecoglou stated, ‘I will never talk to the referee about the rules of the game. I think it is so hard for referees to officiate’ (Article 10). The difficulty mainly comes from an ‘overly complicated rulebook’ (Article 2), the quantity of decisions needed to be made, the level of subjectivity needed for each decision, the speed in which the decision must be made, and the pressure in making these decisions. Exploring this complexity, these factors are summarised in a long-read article exploring the world of Premier League referees. It noted ‘The game only has 17 laws, but the rulebook runs to more than 100 pages. Today’s Premier League referees have to make about 300 decisions per game. (A “decision” is classed as every time a referee could blow their whistle, not every time they do. Many decisions go unnoticed, because the referee opts not to stop play.) Some decisions concern matters of fact, such as awarding throw-ins when the ball goes out of play. But many decisions are interpretations of physical coming-togethers, and a small percentage of these have the potential to shape the outcome of the match. Football is low-scoring, which means the referee’s role is particularly crucial; a single call can swing a game’ (Article 37). Further highlighting the role interpretation plays in making a decision, one journalist describes football as a “sport of a billion subjectivities, from whether an attacker was interfering with play to whether a defender’s hand was in a natural position (Article 9). In addition, ‘these are all factors a referee has to take into consideration, in pretty much a split-second, before making a call that could drive any number of ex-pro pundits doolally’ (Article 28). However, ‘the objective standard of proof for which football’s lawmakers so gallantly strive does not, in fact, exist’ (Article 9). As Neil Swarbrick, PGMOL’s VAR Manager noted ‘all we can strive for is consistency across a single game’ (Article 37), which is in contrast to fans who want ‘decisions that are both objectively correct, and for your team not to be on the receiving end of them but you want it delivered instantaneously’(Article 9).

Culture of referee grievance

The relationship with the referee goes beyond a lack of empathy and depicts a culture of referee grievance. As one journalist noted, “we don’t love referees, so our openness to forgive them for erring is far less than it is for players (Article 16). ‘On the pitch, referees are routinely booed, insulted, told to fuck off. Most referees have stories of crowd members throwing coins or bottles at them, of being spat at’ (Article 37). This is not just for profession football, ‘to be a volunteer match official is be exposed now to routine toxic abuse’ (Article 36). ‘Officials are constantly harangued by players, by managers, by the crowd’ (Article 16). One journalist highlighted, ‘Mistrust of authority, questioning those in power, scrutiny of the decision-making process: these are actually the signs of an engaged public’ (Article 9). However, it is often ‘grotesquely excessive’ (Article 9) and rare occasions turns violent/criminal. ‘In 2021, after Mike Dean sent off a West Ham player, his daughter received death threats online. “They were saying they knew where we lived and they were going to petrol bomb the house”, Dean told the BBC’ (Article 37). Indeed, ‘this behaviour is not harmless and has real-world effects, including the widespread and often violent abuse of referees at grassroots level, cannot be ignored’ (Article 21). Furthermore, it was this type of behaviour that prompted the need for VAR in the first place. The family of Referee Michael Oliver “received death threats after he gave a late penalty to Real Madrid and sent off Gianluigi Buffon of Juventus in a Champions League quarter-final in 2018. Afterwards, the Juventus president Andrea Agnelli used the game as irrefutable proof that VAR must be introduced ‘as soon as possible’, accusing Uefa of ‘scientifically’ damaging Italian clubs (Article 9). Interestingly, one journalist posited the ‘possibility that the best solution to improving refereeing standards is to stop demonising individuals and castigating their decisions which has never been tried’ (Article 21). However, the ‘call for supporters and pundits to remember that officials are human beings with challenging jobs and that mistakes are inevitable would seem like howling into the wind at this point’ (Article 28). However, in the absence of this empathy the level of tolerance of referees and mistakes are very low. ‘’People don’t expect a striker to score every time they shoot, but for us every decision needs to be correct’, he [Darren England – Referee] said. No one notices the things referees do right, but everyone remembers the errors’ (Article 37).

Socio-technical impact of VAR implementation

More accurate but more complicated

With aim of VAR to support the decisions of officials ‘it’s worth noting that it correctly overturned 42 incidents in the first half of the season’ (Article 37) and ‘the number of “correct” decisions in this season’s Premier League has actually gone up compared with last season’ (Article 9), which had ‘116 overturns’ (Article 16). However, implementing VAR required a VAR protocol to be added to the rules of the game. Now ‘the laws are more complex’ (Article 37). Released VAR audio communications has revealed ‘people scrambling to enact a complicated process, to make the right calls and to do so as quickly as possible, all the while having to listen to the referee, on his open mic, wrangling gaggles of furious players’ (Article 34). The game now has fouls that are perceived as being new like a ‘micro-offside’, ‘a slow-motion red card’ (Article 27) and has created a perception of ‘overinterpreting the rules’ (Article 12). A controversial red card given by a VAR intervention was described as ‘a very modern kind of red card: a red card that wasn’t obviously unfair or incorrect; but which also wasn’t obviously correct or fair. It was a red card that happened because the tools exist to call it into being’ (Article 27). Listing to audio feeds of VAR in operation ‘sounds complicated, pressurised, thankless. It has also made the actual referee’s job even harder. So many plates to spin’ (Article 13). ‘All the people involved – on-field ref, linesman, fourth official and especially now in this case VAR’ (Article 13), are ‘working constructively together under intense pressure’ (Article 34). Furthermore, contrary to expectations, people ‘fully accept the pressures that match officials work under, but these pressures are supposed to be alleviated, not exacerbated, by the existence and implementation of VAR’ (Article 24).

Longer and less enjoyable (live)

- Due to VAR, certain ‘decisions now happen long after the action has taken place rather than in the moment’ (Article 1). VAR has been described as being ‘agonisingly slow’ (Article 37) with the result of making the ‘game too slow’ (Article 10). Such delays have been highly criticised, ‘a pause of even a minute, while the VAR deliberates, can feel endless. (In its first season, VAR once took 3 min 45 sec to rule out a goal by Sheffield United.) And the longer it takes, the more pressure there is to get it right’ (Article 37). However, as noted by one journalist ‘precision takes time’ (Article 37) and the PGMOL has emphasised ‘efficiency should never be prioritised over accuracy’ (Article 19). Nonetheless, there is strong evidence to show that while the decisions are more accurate ‘VAR is also changing the game, critics say, for the worse’ (Article 1). Indeed ‘a survey by the Football Supporters’ Association revealed that 95% of fans – be they at home or in the stands – found video refereeing technology made their experience of watching a game less enjoyable in 2021. Only 26% were in favour of keeping VAR’ (Article 47). For match going fans especially, ‘for them it’s not so much the issue of interference or inconsistency as the time taken out of matches for VAR to do its work, its the throttling of spontaneity’ (Article 47).

Negative emotions

- Given the hedonic nature in which professional football is consumed, emotion plays a significant role. As noted by one journalist ‘At its most enjoyable, football is fast and charged with emotion’ but ‘ever since it was introduced, VAR has been making people furious’ (Article 37). The emotional impact for fans was summed up by one analyst who noted that fans may feel ‘VAR has killed the emotion of football, and yet for some reason you seem to feel more strongly about it than most other things in your life’ (Article 9). This emotional effect is not only felt by fans but also managers. As manager Pep Guardiola stated, ‘emotions for the manager after the games is difficult, the feelings are so, so high and it’s difficult to handle it’ (Article 11). Other emotions and feelings expressed by managers include ‘sick’ and ‘embarrassed’ – manager Mikel Arteta (Article 11), ‘amazed’ manager David Moyes (Article 43). Pundits have also vocalised their emotions as ‘disgusting’ – Alan Shearer (Article 2). Jermain Jenas posted an emotion filled tweet after a VAR intervention in a game with his previous club. The tweet reference officials as ‘ruining the game!’ but Jenas later apologised saying ‘emotions get the better of me’ (Article 28).

Self-fuelling form of reality entertainment

- Furthermore, ‘Sky Sports is only inflaming the situation, much like all the other TV and radio shows further stirring the pot and winding up already incensed fans by zeroing in on errors. You can’t blame them, as it makes for better viewing and is way easier than analysing the countless in-match mistakes made by managers and players. Meanwhile on the phone-ins, the ranters and ravers will continue to get their radge on, screaming the odds as they pour more petrol on the flames’ (Article 110). In addition, ‘VAR has just provided fans who take these things far too seriously a new layer of officialdom at which to vent their spittle-flecked fury’ (Article 110). Indeed, the extra review and analysis of decisions that VAR has brought about only adds to the emotion, but some have said ‘eventually it’ll become more popular than the football’ (Article 28). As one journalist noted ‘we’re always hearing about the need for content to pad out those aching, hours-long gaps between matches. The Real Match Officials Of Stockley Park might just be it’ (Article 34). ‘In a world where the TV discussions before, during and after Premier League games invariably focus more on refereeing decisions than actual football, fan anger at real or perceived injustices by on-field officials and their screen-watching assistants will naturally fester. The inevitable consequence will be even more scrutiny: rinse, recycle, repeat’ (Article 46).

Referee authority erosion

Examining the impact that VAR has on people, it is possible to identify a change in the relationship dynamics between (i) the officials and all the other stakeholders and (ii) between officials themselves. This is specifically explored in one article which stated ‘VAR has affected the relationship between players and officials in more intangible ways. We change things by observing them – and everything is observed now, from players and officials, to officials observing players, to officials observing the observations of other officials’ (Article 36). More specifically, the introduction of the VAR and AVAR has made referees look weak with their power being eroded. This sentiment is coming particularly from managers such as Ange Postecoglou who stated that ‘Their authority is constantly getting diminished. I grew up afraid of referees. They were like policemen’ (Article 10). It also comes when a referee’s decision is overturned by VAR as indicated by manager David Moyes ‘the referee should have stuck to his own guns. The sad thing is this is the level of the weak refereeing at the moment’ (Article 43). Indeed, the overturning of decisions gives a perception of the referee being a ‘fallible middleman’ (Article 36), with VAR’s role increased to officiating the match, not supporting the on-field referee (Article 21). It must be noted that one manager Pep Guardiola highlighted that referees and VARs act like movie stars and should ‘be more humble’ (Article 19). This perceived weakness is the opposite of what VAR was intended to do, nonetheless, it has created ‘a collision of competing versions of the truth’ (Article 36), where players and managers can apply additional pressure on referees to gain a competitive edge.

Low transparency

The VAR protocol outlines that ‘broadcasting the conversations between on-field officials and VARs live is currently prohibited’ (Article 13). As a result, confusion is created and a lack of insight into officiating decisions. An example of the confusion is captured in a description of a disallowed goal for Arsenal due to an VAR offside decision, however, during the VAR check a foul was picked up and a penalty given in place of the disallowed goal, which in total took four minutes and 39 seconds. ‘On the pitch, in the stands, in households around the country and even in the BT Sport commentary box – nobody except the referee, Craig Pawson, and his assistants seemed quite sure what exactly was going on’ (Article 45).

Lower tolerance for augmented errors

With VAR, officiating is no longer conducted through the eyes of the referee, as their ‘expert judgment [is] being augmented by an off-pitch official viewing match footage’ (Article 1). Refereeing is now in ‘ultra-high definition’ (Article 25). As a result, the officiating failures because of a ‘human error’ are no longer acceptable to vested stakeholders (Article 38, Article 24). Human errors are hard to accept as these were ‘just what VAR was supposed to prevent but cannot as it is a very human process’ (Article 35). People have even criticised one referee for using the word ‘mate’ during a VAR audio recording (Article 32), even though such language is used for personal support during the decision process. As one journalist noted, ‘create a word cloud of the VAR exchanges and the most prominent term would be “pal”’ (Article 34). To counteract this trend, Howard Webb has started to publically air these VAR communications ‘as an exercise in humanising the most commonly demonised figures’ (Article 34).

Increased subjectivity and controversy

Notwithstanding the view that fan ‘expectations are unduly high’ (Article 28), VAR was ‘was brought in to reduce the number of controversial decisions … [but] many fans, with good reason, think the opposite’. A key issue with VAR is that it cannot account for the level of subjectivity in officiating decisions. ‘once there is subjectivity, there will never be perfect consistency’ (Article 37). As a result, ‘controversy is and always will be part of sports, no matter the technology’ (Article 12). Roberto Rosetti, the Chief Refereeing Officer at Uefa noted ‘perfection in football doesn’t exist…. technology works well for factual decisions. With Hawk-Eye in tennis, the ball is in or out. It cannot be maybe in. But football is a game of physical contacts. There is always some kind of subjective evaluation, and that’s why it’s so difficult’ (Article 37). Indeed, having additional eyes on a game means additional opinions as ‘different referees will judge incidents differently.for many decisions, “there is no correct”’ (Article 37). In addition, VAR introduced another layer of subjectivity, which is the ‘clear and obvious’ grounds for intervention. That is, VAR needs to determine if a clear and obvious mistake has been made before making an intervention. Indeed, ‘this is one of the reasons it takes time to make an overturn call, the VAR has to be as sure as they can be that an error has been made’ (Article 28). Furthermore, Richard Bevan Chief Executive of the League Managers Association (LMA), has called for ‘a review (and simplification) of the interpretation of the term “clear and obvious” in VAR decision-making, as this is a cause of much confusion at present’ (Article 7). The confusion is not only with managers, referees also struggle to agree on what is clear and obvious. Showing a clip of an incident in a room with ‘some of the best referees in world football, with decades of experience, and even after seeing the incident multiple times from multiple angles, they couldn’t agree about what was clear and obvious’ (Article 37). Indeed, this creates the scenario where controversy is created when a VAR intervenes and also when it doesn’t for a decision that ‘should have been ruled out by VAR but wasn’t’ (Article 46).

Use of latent technology potential

VAR has a latent technology potential to pick up much more infringements in Stockley Park than the referee on the field. However, ‘just because the technology exists, there’s no need to go looking for offences’ (Article 33). Roberto Rosetti (Chief Refereeing Officer at Uefa) highlighted that too often VAR is being used to ‘investigate every single detail’ of matches (Article 37). ‘Once, when Rosetti experimented with using VAR to review every incident in a single match, he found seven penalties and three red cards, according to a strict reading of the laws of the game’ (Article 37). This can create a scenario where ‘every decision gets forensically checked and we will be sitting around for a long time in every game trying to figure out what is going on’ (Article 10). As Rosetti noted, the problem ‘lies not in the technology itself, but in how it’s being implemented’ (Article 37).

Continual change

The implementation of VAR was predicted to take time ‘The head of PGMOL, Mike Riley, has always argued it would be a five-year process to get things right (though he didn’t mention things getting worse before they got better)’ (Article 47). As such, change is to be expected, as with any implementation. The result of these continual changes is the perception of an ‘unplanned experiment, bruised and buffeted by those strange new pressures, unintended consequences of a technology that nobody seems to be in control of right now’ (Article 36). Nonetheless, the changes have been positive as well as negative.

Technology calibration

These are changes that calibrate the technology to create a better fit with the decision process. A good example of this is change in the thickness of the offside line in. ‘The lines used to determine offsides had been revolutionised (translation: made slightly thicker) in order to put an end to ridiculous decisions determined by rogue toenails and armpit’ (Article 46). Mike Riley, Managing Director of the PGMOL (at the time), stated the change was predicted to give ‘back 20 goals to the game that were deemed offside’ in the previous season (Article 46). This eliminated decision such as the ‘micro-offside’ (Article 27), where players can be judged be offside by a matter of millimetres.

Rule calibration

In an effort to be more objective, more consistent and aligned with VAR, the handball rule was modified. In simple terms, the change highlighted that ‘if the silhouette of the body is deemed to be “unnaturally big” then a penalty can be given, whether the player meant to touch the ball or not’ (Article 2). This result has been a lot of confusion. Manager Ange Postecoglou said, ‘I’ve got no idea about the handball rule’, he said. ‘I really don’t. It just seems if it hits your hand, it’s a penalty, and then other times if it hits your hand, it isn’t a penalty. It is the one rule in the game I just don’t understand. Unless we start developing armless defenders, I don’t know how you are supposed to block things and be in a natural position’ (Article 28). ‘But VAR loves this because a ball striking the arm is less subjective than a shoulder-to-shoulder tussle or a tackle that might be fractionally mistimed or over-aggressive. There is the ball, there is the arm, there is contact between the two, therefore referees can do something’ (Article 33). However, as well as legitimate challenges, there has been ‘constant incorrect criticism to the continual law changes’, which may inflate the impact of such changes (Article 29).

VAR protocol calibration

From the use of VAR changes have been put in place because of gaps in the protocol itself which have resulted in rare but highly ‘embarrassing’ (Article 19, 26, 28) or ‘systemic failures’ (Article 19). The most publicised has been a miscommunication error between the VAR and referee that disallowed a valid goal in a game between Tottenham Hotspur and Liverpool. When England said, ‘check complete’, using the official term to show a VAR has endorsed a referee’s on-field verdict, he confirmed an erroneous decision. The VAR (Darren England) ‘thought the referee, Simon Hooper, had given a goal to Luis Díaz, but he hadn’t been paying attention and the goal had in fact been disallowed for offside. When England said, “check complete”, therefore, using the official term to show a VAR has endorsed a referee’s on-field verdict, he confirmed an erroneous decision’ (Article 21). The error was compounded by the existing protocol which prohibits a decision change once play has restarted. Since that event Webb (Chief Referees Officer, PGMOL) highlighted that ‘a lot of steps in place to ensure the error we saw in that important game doesn’t happen again’ (Article 14), resulting with ‘in-game communication protocols being altered’ (Article 7).

Discussion

Exploring the socio-technical dynamics of VAR, it is possible to map an interconnected web of impacts, as a complex socio-technical system, that has arisen because of the technology. In a number of cases, these impacts seem incongruent (e.g. more accurate but less enjoyable games), which in itself is a key source of confusion and dissatisfaction with the system. One potential explanation is that from a utilitarian perspective represented by theories such as TAM (Venkatesh et al., 2003) or the Information Success Model (DeLone & McLean, 2003), VAR has been a success, but from a hedonistic perspective represented by the Hedonic-motivation system adoption model (Lowry et al., 2013) it has been a failure. Our analysis suggests that while decision accuracy has increased and errors have been reduced, fans, players, managers, and pundits feel the technology is taking the enjoyment out of the game. Yet, while it has made the job of the actual user (officials) more difficult, they have not requested it to be removed, which is in direct contrast to the non-officiating stakeholders and governing bodies. As a result, a wider framework or theory for IS success is needed to fully understand the multi-stakeholder system that goes beyond the user and addresses the broader cultural impacts and complexities that includes a new form of entertainment which inspects VAR decisions on TV and social media, which further triggers more emotions, and puts more pressure on referees.

From the dataset analysed there is evidence of error cascading throughout its implementation (Renger et al., 2017). Indeed, the motivation of VAR implementation was to eliminate errors, yet it seems that while it has removed errors, it has also introduced others. Furthermore, in an effort to mitigate these new errors, other errors have appeared. A lot of these can be characterised as unintended consequences (Baert, 1991), yet the action on their solution seems to create more unintended consequences. In a sense, the result is a cascade of unintended consequences. For instance, by creating more efficiency, the game is now longer, resulting in fans putting pressure on officials to make faster decisions. In an attempt to make faster decisions, other errors occur, like the one caused by rushed communication between the VAR and the referee, which essentially disallowed a valid goal in September 2023 (Tottenham Hotspur v Liverpool). The impact of this error compounded the lack of perceived effectiveness of VAR, which continues to erode public trust in the system and creates an even lower tolerance for errors.

The use of latent technology highlights the difficulty in refraining from using the full potential of the technology, even though the key requirement for that technology is quite low. The result is a more nuanced approach to utilising the technology that aligns with the spirit of the game but not necessarily to the letter of the law and adheres to the authority of the referee. However, this highlights the increasing amount of subjectivity being introduced, which in-turn creates additional issues. This finding closely parallels the role of technology in overdiagnosis and overtreatment (Hofmann, 2015). It is argued that technology can create a self-perpetuating loop of diagnostics where even milder cases are treated as well as cases that would otherwise not have been detected (Hofmann, 2015). As a result, we get a technological construction of disease very similar to the VAR-constructed fouls and red cards.

Concluding remarks and theorising outputs

The underlying contribution of this study is that it provides a basis to explore multiple IS and socio-technical phenomena that are present in the well-documented VAR implementation and use. Our grounded approach provides the first step in capturing the moving parts in the VAR implementation and use, in the form of the 14 categories, organised as ‘Socio-Technical Context’ and ‘Socio-Technical Impact’ (see Figure 2). These initially point to a need for a multidimensional theoretical frame to fully understand what is happening with the VAR implementation and use through exploring the relationships between these categories.

Figure 2. The 14 categories across the ‘socio-technical context’ and ‘socio-technical impact’.

To conclude, VAR is not just a technology but is part of a complex socio-technical system. Its effectiveness and impact are shaped by the existing social structures of football, the agency of individuals involved, and the continuous appropriation and adaptation of the technology in response to various experiences and feedback. This analysis underscores the importance of considering both the technical capabilities and the social context in which technology is deployed, especially in high-stakes, high-visibility environments like professional sports.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. https://www.fifa.com/technical/football-technology/standards/video-assistant-referee