There have been several high-profile controversies in football over the years where human error has had a substantial influence on match results—from Diego Maradona’s “hand of God” goal against England in 1986 to Frank Lampard’s disallowed-despite-the-ball-having-clearly-crossed-the-line goal against Germany at the 2010 World Cup. Inevitably, such incidents left supporters feeling outraged and let down, sparking calls for a solution.

Goal decision systems and video referees could easily have corrected such errors in judgement by their on-field counterparts, yet we find ourselves over a decade down the line from that phantom Lampard goal, with goal-line technology and VAR widely used in European domestic leagues and at major international competitions, debating whether its impact on the game is a positive or negative one.

VAR has proven to be a divisive subject among footballers, fans and pundits alike. Despite the intention being to correct "clear and obvious errors" and "serious missed incidents", the intervention of technology appears to have split opinion. Some argue that human error is part of the game and that technological interference ruins the flow of play, whereas others believe that removing the possibility of misunderstanding and misinterpretation influencing match results can only be a good thing.

Offside decisions
The skills and attributes required to be match official should not be underestimated, particularly when it comes to making visually demanding judgements on the pitch. With offside decisions often being especially challenging due to speed of play and line of sight, referees and their assistants are faced with the challenge of tracking the movements of at least four different moving objects at once (two defenders, the attacker in question, and the ball), all while other players are moving around in their field of view.

Even if all relevant players and the ball are grouped closely together, they have a very short window of time to process the information in front of them and make a call. Those refereeing at a high level, however, have developed visual strategies to overcome such challenges, and instead fixate on the offside line whilst using their peripheral awareness to determine whether a player is offside. These quick but informed decisions minimise interruption to the flow of the game, allowing play to continue without disruption.

Offside decisions influenced by VAR aren’t immune to error or interpretation either, and remain a major area of dispute and debate due to the fine margins under which the technology operates. VAR reviews involve projecting a one-pixel wide line onto the TV image of the pitch at the exact positions of the parts of the attacking and defending players that can be used to score goals. However, evidence in vision science points to problems with this process. Broadcast cameras operate at 50 frames per second, with each pixel in the TV image covering a distance on the pitch of around an inch (2.5 cm). A still-frame from that footage, however, includes players and the ball itself whilst in motion, resulting in the point in question being smeared across roughly 8 pixels. Those 8 pixels equate to approximately 8 inches (20 cm) on the pitch itself, with the point of contact with the ball being located somewhere within that blur, meaning that there is a margin of error of 4 inches (10 cm) either side of the true position. That’s fine when it comes to clear offsides where they are off by 4 inches or more, but highly skilled players operate at the finest of margins, making close calls subject to incorrect decisions as a result of the 8-pixel wide fog.

Is VAR any more reliable than on-pitch officials?
Alongside a high level of physical fitness and an exhaustive knowledge of football rules and regulations, it stands to reason that good eyesight is an incredibly important attribute of a great referee. A 2011 study analysing the decision-making accuracy of expert referees, found that those who made more successful decisions had better visual skills, including peripheral vision and visual reaction times, highlighting just how important vision is to being a top-class referee. That makes it all-the-more surprising to learn that a 2017 study found that 18% of top-level officials from the Portuguese football league, many of whom also officiate at international level, had not had an eye test in the past three years, if at all, seemingly making shouts of “are you blind, ref?!” more of a legitimate question than a cruel taunt.

However, according to the body which represents Premier League referees, PGMO, Premier League referees make an average of 245 decisions per game, three times more than an average player touches the ball over the course of a match. That's one decision every 22 seconds. Approximately 45 of these decisions are technical (goal-kicks, corners, throw-ins etc), with the remaining 200 relating to physical contact and disciplinary actions. Of those 200, around 35 are visible decisions where an action is taken (fouls, restarts etc) and 165 are non-visible, where play is allowed to continue. In total, referees make around five errors per game, meaning they are right 98% of the time.

Comparatively, the International Football Association Board found that, in the 804 matches analysed during VAR’s two-year worldwide trial, there were 3,947 checks for possible reviewable incidents, with 56.9% of checks being for penalty incidents and goals, and that the VAR system was accurate in 98.9% of decisions.

What do you think of the use of computer vision technology in football? Is the additional 0.9% of accuracy worth it or does VAR detract from the magic of the game for you? We’d love to hear what you think.

Sources:

1. Wired.
2. Sage Pub.
3. Performance Vision.
4. Sport Performance Analysis.
5. NCBI.
6. PubMed.
7. InnerDrive.
8. BBC.