Regular-Season Dossier: The Laws Are Clear, the Referee Remains the Variable

In the last ten Bundesliga matches I rewatched minute by minute, 32 substitutions fell between the 60th and 75th minute — an average of 3.2 per match. That rhythm does not belong to tactics. It belongs to the clock, and the clock is the referee's territory.

When a team is granted five substitutions instead of three, the second half is no longer an unbroken block of 45 minutes. It is cut into short segments, each corresponding to a change, a quick touchline huddle, a fourth official raising the electronic board. Inside each of those cuts, the referee is pushed into a new grey zone: stop play or let it run, count the time or not, add how many minutes for a stoppage he himself just permitted.

A regular season has no final to draw every eye to a single point. There is no knockout round forcing every error into the open inside 90 minutes. Errors here flow slowly, accumulate round by round, and by the seventh month of the season they have become a trend. A trend can be measured.

I follow matches with a spreadsheet open on my second screen: substitution minutes, actual ball-in-play time, number of VAR interventions, stoppage time displayed and stoppage time actually played. This method began with a mistake of my own. In 2026, while an intern in Shenzhen, I misnamed a striker four times in the first half of the opening match of the Under-20 World Cup and was corrected bluntly by my editor. For the next two weeks I sat through the entire group stage just to record the correct names and shirt numbers of 120 players. The habit of standardising data before writing was born there, and it applies to referees as well.

The current Laws of the Game are not vague on the points spectators argue about most. Law 3 sets out clearly the number of substitutions and the cases of substitutions outside the list. Law 11 defines offside clearly. Law 12 defines handball and unsporting behaviour clearly. IFAB publishes the text, amendments carry dates, and member associations apply a common framework. The problem lies elsewhere: the same sentence can be read by two referees into two different thresholds of application, and neither of them violates the text.

The biggest variable of a regular season sits in the threshold of application, not in the Law itself.

Three changes shape the context of this season. The temporary measure allowing five substitutions — originally introduced during the pandemic to protect player fitness under a compressed calendar — has become the norm in most competitions. Some leagues still keep three, and that inconsistency produces two different sports in terms of match management, under one and the same Law. Competitions have simultaneously tightened time-wasting, including a rule limiting goalkeepers to eight seconds in possession, with a corner kick awarded to the opposition as the sanction. And semi-automated offside has expanded, cutting the determination time to a few dozen seconds while shifting the centre of controversy to another layer.

Start with handball, because that is where the gap between the Law and the person blowing the whistle shows most clearly. Law 12 sets out the factors: the direction of the ball, the distance between the point of contact and the body, the position of the arm relative to the body's natural silhouette, and whether the ball touched the hand and then went into the goal or created a goalscoring opportunity immediately afterwards. Four factors, none of which is prioritised in the text. Which means the referee must build an order of priority in his own head, and that order changes from person to person, competition to competition, round to round.

History shows that threshold has shifted many times. In 2026, Diego Maradona's goal against England, known as the "Hand of God", stood, because the Law at the time allowed no form of intervention once the referee had decided. In 2026, Thierry Henry's handball in the World Cup qualifying play-off against Ireland led to a goal and a controversy that lasted years, along with a wave of demands to bring technology into football. In 2026, Luis Suárez handled the ball on the goal line in the World Cup quarter-final against Ghana, was sent off, and Ghana missed the penalty — a situation where the Law permitted the maximum sanction, and that sanction still was not enough to offset the damage. Three events, three ways the Law responded, and all three led to a later change.

When VAR arrived, the widespread expectation was that controversy would shrink. The opposite happened. VAR does not make controversy disappear; it moves controversy from the stands into a room with screens. Previously, spectators argued over whether the referee had seen it. Now they argue over whether the referee applied the right threshold — and that is a far harder argument, because it cannot be settled by a single frame.

VAR's intervention principle rests on the phrase "clear and obvious error" or "missed serious incident". There is no technical definition of "clear". There is no millimetre threshold for "obvious". Which means each VAR team builds its own intervention threshold, and that threshold shifts match by match, round by round, according to the public pressure of the previous week. After a round criticised for too much intervention, VAR teams tend to intervene less in the following round. This is measurable behaviour if you log interventions week by week.

With semi-automated offside, the mechanism is even clearer. The system determines the offside line through tracking data and the point of contact, so the question "was the player offside" is now almost definitively answered. But the offside Law does not stop at position. It also asks whether a player in an offside position became involved in active play: did he touch the ball, did he impede an opponent, did he gain an advantage from that position. That is a judgement, not a measurement. Semi-automated offside solves the data part and leaves the judgement part entirely intact. Controversy does not vanish; it changes layer.

The same situation, two ways of blowing the whistle — the Law is never vague, only the person holding the whistle is vague.

I have rewatched one recurring pattern many times this season: a high defensive line, an offside trap set at the halfway line, a through ball played at exactly the right moment. On the television frame, the distance between two defenders and the striker is sometimes half a step. An assistant referee cannot judge that by eye at that distance. The referee, standing twenty metres from the point of contact, certainly cannot.

A high defensive line is a wager; I only record the moment the gambler turns over the card. A team pushing up trades the space behind it for pressure higher up. When the trap springs at the right moment, they get a counterattack. When the trap springs a beat too late, they concede a direct free kick or a penalty, and the referee is the one who signs the report. There is nothing new tactically. What is notable lies elsewhere: the frequency of such situations in the second half rises markedly once a team has used three or four substitutions to push its defensive line higher.

This is where the substitution Law and the offside Law meet, and it is the place few people look. The fourth and fifth substitutions do not merely bring fresh players onto the pitch. They change the risk structure of the match. A team with five substitutions can sustain high pressing intensity throughout the second half without paying for it in fitness. The consequence is that the high defensive line is held longer, more offside traps are set, and the number of situations depending on an assistant's judgement — human judgement — rises with it. Technology does not reduce the number of hard decisions; it only ensures those hard decisions are made faster and in front of more cameras.

The referee's positioning is an undervalued variable. The classic diagonal system requires the referee to follow a diagonal from one corner to the opposite one so as to always hold an open angle. But when the match shifts into a high tempo with a high line and constant through balls, the referee is forced to choose: stay tight to the last line to judge offside, or drop back to cover contact. He cannot do both. Every choice is a surrender of one category of incident, and the category surrendered will show up in the post-match error review.

Discipline is not there to punish, but so that the match can continue.

The eight-second rule for goalkeepers is the cleanest example of that philosophy. Previously the Law allowed a goalkeeper to hold the ball in his hands for a maximum of six seconds, but the sanction was an indirect free kick at the point of the offence — a sanction almost never enforced, because no referee wants to create a goalscoring situation out of an administrative technicality. The Law existed but did not operate. The new rule replaced the sanction with a corner kick, putting the punishment at the right level: painful enough that goalkeepers must comply, not heavy enough to decide the match.

That handling deserves to be read as a principle. A Law that is not enforced is not a Law; it is a sentence. When IFAB chose to change the sanction rather than the number, it acknowledged something the refereeing world had long known: the problem with time-wasting was never duration, but motive. A goalkeeper holds the ball long because his team is winning and needs time. The sanction has to act on the incentive, not on the clock.

But a side effect appeared here that I have tracked in recent rounds. As the sanction grew heavier, goalkeepers switched to another option: early long distribution, sometimes regardless of team structure. The ball goes forward while the defensive line has not yet pushed up, producing turnovers in midfield and counterattacks that the referee then has to manage. No rule change is neutral. Every change pushes behaviour somewhere else, and that somewhere else requires a different set of refereeing skills.

The rule that only the captain may approach the referee during explanations of decisions is a small change in wording and a large one in match management. Previously, a controversial decision tended to draw a cluster of five or six players around the referee. The time lost there was not fully added back, and psychological pressure on the referee rose at precisely the moment he needed to be calmest. Once the rule was tightened, the cluster vanished from the frame. But the pressure did not vanish; it moved to the captain, who must now both relay information and absorb the frustration of the whole team.

Stoppage time has become a political calculation. I record two figures per match: the stoppage minutes displayed on the fourth official's board, and the minutes the ball actually rolled inside that window. The gap between the two is usually small, but how it is produced is notable. Most stoppage time does not come from injuries or VAR, but from administrative stoppages: free kicks, throw-ins, substitutions, celebrations. Those are stoppages the referee has the right to add, and also stoppages he can choose not to add if he judges a team is abusing them.

That choice is what produces inconsistency. Two matches with the same number of stoppages can differ by three or four stoppage minutes, depending on whether the referee reads the situation as deliberate time-wasting or genuine injury. No device measures intent. Only the referee.

Average stoppage time in top leagues has risen considerably in recent seasons. The cause is not that matches contain more stoppages, but that the instructions for calculating stoppage time have changed: administrative stoppages must be fully added. The consequence is that a match can run close to 100 minutes of ball-in-play time, and within that added period the quality of the referee's decisions declines through fatigue. This is a trade-off rarely discussed. We demand more playing time, and we pay for it with decisions in the 95th minute made by a man who has already run more than twelve kilometres.

I believe in the naked eye, but VAR taught me that the naked eye also lies.

This is the lesson I learned latest, and it changed how I watch matches. I used to believe that an experienced referee, standing in the right position, would see correctly. After years of rewatching incidents at frame-by-frame speed, I understood that the human eye does not record events; it records memories of events, and memory is bent by speed and angle. A challenge at real speed looks like a stamp. The same challenge at frame speed looks like a light touch. Both are true, and the referee is permitted to see only one of them. That is why I no longer argue with decisions on television. I argue with the process that produced the decision.

The first mistake is not there to be erased, but to be compared against later.

This is the principle I apply to both my writing and my watching. Every controversial incident should be logged with its context: round, minute, score, referee, and the pre-match instructions the competition organiser sent to the clubs. Three months later, when a similar situation occurs and is whistled differently, the old data becomes the measuring stick. Without it, every debate restarts from zero, and every referee can justify himself with the phrase "it depends on the situation".

A regular season is the ideal environment for this kind of comparison, because the sample of matches is large enough to form patterns. A single decision says nothing. Thirty decisions of the same type across thirty rounds say a great deal, and most of what they say is not in the text of the Law.

In my daily work I read a great many disciplinary rulings. One thing is easy to notice: disciplinary panels operate almost entirely on precedent. A sanction is issued not only on the basis of the act, but on the basis of similar sanctions in the past. When there is no precedent, they create a new one, and that precedent binds them in the following season. This mechanism works well at the administrative level. It cannot work at pitch level, because a referee has no time to consult precedent in two seconds. This is the structural gap between the disciplinary system and the match-management system, and it cannot be filled with technology.

There is an assumption underlying almost the entire debate about refereeing: that inconsistency is a fault. I would argue that in some cases, inconsistency is the inevitable consequence of a system designed to adapt, and that demanding absolute consistency would produce a worse kind of football.

Take the classic situation. A defender lets the ball touch his hand inside the penalty area in the 12th minute and in the 88th minute. Same player, same arm position, same distance. By pure logic, the two situations must receive the same decision. But a referee is not running an equation; he is running a match, and the match in the 88th minute is in a completely different state. If he awards the penalty in the 12th minute, the match still has 78 minutes to self-correct. If he awards it in the 88th, he has almost decided the result. That pressure is not in the Law, but it is in the referee's head, and no training course erases it.

This is the biggest blind spot in every public debate about VAR. Fans demand consistency as a technical standard. Referees are required to deliver consistency as a management standard. The two are not the same thing.

I am not excusing this. I am only noting that if the standard were "every identical situation must be whistled identically", the consequence would be that referees award more penalties in the 12th minute and fewer in the 88th — meaning they are still applying a different threshold, just hiding it behind a different reason. Formal consistency does not produce substantive fairness. It only makes inconsistency harder to detect.

The only way out of this loop is to make the threshold transparent rather than conceal it. If a competition announced that handball situations in the penalty area would be assessed under a specific order of priority, and that order of priority applied across the whole season, then at least spectators would know what they are watching. At present, the order of priority exists, but it lives inside referees' heads and inside closed meetings before each round.

The trend of the regular season is heading in a clear direction: referees are increasingly judged less on whether they were right or wrong, and increasingly on whether they were consistent with themselves across the season. That is a reasonable shift, but it only has value if the data on that consistency is made public. A referee can be consistently wrong across thirty rounds, and if nobody measures it, nobody knows.

My proposal is concrete: every competition should publish, at the end of the season, a compilation of controversial incidents with the final decision and the instruction applied. No need to name referees. Just the situation, the threshold, and the outcome. Three seasons of that would build a data set thick enough for anyone to verify for themselves instead of arguing on feeling.

The Law is clear. What remains is to record it properly.