Football substitution is an issue I am currently researching. Although the sources speak a clear language. But there have been many exceptions in football substitution, as match reports show.
I know of exceptions in some countries on the European mainland between the First and Second World Wars. The four British associations also knew about this and tried to press FIFA to comply with the rules first. But in the end (1931) they unofficially let the continental European way continue. It had already become a common law there in the few years it had been in existence, and it was much appreciated.
In this post I give an overview of the development of football substitution as far as I can draw/describe it at the moment.
Football substitution in the 19th century: Emergency!
In the second half of the 19th century, football was not yet so widespread outside the UK. And substitutions in our modern understanding did not exist, but there was another habit that had the same name. If not all the players of a team were present at the start of a match, the missing player was allowed – by prior, mutual agreement – to be replaced by another person for the duration of his or her delay.
However, there was rarely a substitute player in the team, which is why a spectator was accepted as a player instead. However, s*he was not always recorded with her*his name, but sometimes as a substitute as S. Ubstitute or E. Mergency. The rules of the Eton Field Game in the 1850s did not call it substitution, but emergency, which describes it much better: It was an emergency that substitution was provided for.
“The Charterhouse evelen played a match in cloisters against some old Carthusians, but in consequence of the non-appearance of some of those who were expected it was necessary to provide three substitutions”
– Bell’s Life in London, an Sporting Chronicle, February 22, 1863. p. 7.
In an international match, there was the case where a national team was replaced by the player of the local football club. On April 15, 1889, an international match between Wales and Scotland took place in Wrexham. The actual Welsh goalkeeper did not arrive – it took 30 minutes for another international goalkeeper of the Welsh to arrive in Wrexham. So for the first 30 minutes a Wrexham amateur footballer took over as goalkeeper. The game ended without a goal.
Half a century later, the first mention of football substitutions can be found in the minutes of The IFAB’s Annual General Meeting. In 1923, substitution is only possible in non-competition matches and in the event of a serious injury, and only after prior agreement between the two teams. The referee must also be informed of the agreement, as must substitutions during the match.
Exceptions in continental Europe between the world wars
I cannot yet say with certainty which countries circumvented the substitution requirement during this period. In 1930, FIFA stated that it had become the custom for some associations to change players at non-competition matches, not only in the event of injury. In 1931, a Dutch official wrote a letter listing the countries that were circumventing the rule (including in competition matches?): Denmark, Germany, Austria, Switzerland, Hungary, the then Czechoslovakia, Italy, France, the Netherlands, Belgium and – outside Europe – the USA. How these countries circumvented the rule was varied: Either only the goalkeeper was allowed to be replaced or generally players only during the first half or only until 20 minutes before the end.
Carl Koppehel, a German referee, functionary and publisher of the German Referee Newspaper (“Deutsche Schiedsrichter-Zeitung”) also names Scandinavia as one of the countries insisting on substitutions in football. Now, Scandinavia is not a country, but a term for several countries. However, Koppehel does not name it more precisely.
Koppehel was also a vehement opponent of football substitution. He described it as unsportsmanlike to allow the uncontrollable substitution of allegedly injured players. In his opinion, it was better to accept a weakening of the team through injury. In contrast, the previously mentioned Dutch official reported that public opinion in the Netherlands considers the prohibition of substitution to be unfair. “An unsportsmanlike action just as well as any abuse of this courtesy”. If a player is unable to play because of a serious injury and a team has to play with one less player for the rest of the season, then there is an imbalance.
As already written in the beginning: The four British associations unofficially tolerated the exceptions allowed by FIFA in 1931, while they still did not allow substitutions. They were of the same opinion as Carl Koppehel.
What if a serious injury is faked?
What if the substitution was only made for tactical reasons?
After all, you can’t look the players in the head!
And therefore it remained completely forbidden.
These are arguments that make us both smile and seem familiar. Today, we’re changing mainly for tactical reasons.
But the argument “can’t look into the players’ heads” is still used today when discussing about foul play. Was a foul play deliberate, i.e. deliberately negligent?
And the justification of a tactical unfairness also came up in spring 2020, when the five substitutions per game were discussed and decided. This temporary change to the substitution rule was introduced to keep the risk of injury low after the week-long break and the only short preparation phase before the restart. But what if these two additional substitutions are used for tactical reasons rather than health concerns?
The same was discussed 90 years earlier: In case of injury, replacing a player is certainly fair, but not to make tactical finesses. After all, football is an honest competition. And tactical gimmicks were looked at sniffily, especially by those who were against substitutions.
After the Second World War: The dogma of football substitution is crumbling in Great Britain
Shortly before the Second World War, it was hardly imagined that a decade later the dogma of football substitution was beginning to crumble. In Nazi Germany in 1938, an injured goalkeeper was only allowed to be substituted in friendly matches – nobody else. And in the UK, it was not only the teams that had to give their mutual consent before friendly matches of international matches, but also the two national associations.
Nine years later, in 1947, the Referees’ Committee advised FIFA to allow the change of goalkeeper always plus the replacement of a field player during the first half. Only in the case of serious injuries, of course.
FIFA submitted this proposal to the Annual General Meeting of The IFAB in 1948, but it did not receive the necessary majority. The same was true of other FIFA proposals, namely the substitution of two injured players up to the 42nd minute (1947) or the substitution of the goalkeeper (only) in competition matches (1948 and 1949).
And although The IFAB rejected FIFA’s proposal in 1948 to allow the replacement of an injured goalkeeper at all times and of an injured fielder during the first half, they had this option tested. Presumably with the approval of the four associations from Great Britain. It was tested in 1951 at a FIFA international youth tournament, in the qualifying for the World Cup (1953) and at the 1954 World Cup itself.
That’s what made it so confusing for me at first: if FIFA’s proposals were rejected in those years, why were there players substituted at the 1954 World Cup and in the qualifying matches?
They were experiments.
And the home nations? After FIFA’s experiments in 1951, 1953 and 1954, FIFA submitted the proposal again in 1956 and 1957 – and it was rejected both times. A clear signal.
However, there was a slight concession in 1957: The IFAB granted the national associations to allow the substitution of injured players. In 1958, the committee added a recommendation that an injured goalkeeper could always be substituted plus one injured field player during the first half.
The reason for this was the increasing number of teams in FA Cup matches, whose injury-related fatigue had a decisive influence on the outcome of the games.
Let’s go: Substitutions independent of an injury
In 1965 there was another innovation in football substitutions, which was not immediately reflected in the Laws of the Game: In 1965, two substitutions were allowed in England, even in competitive matches, irrespective of an injury. In the 1966/67 season, the Scottish FA also adopted this rule. And again, a year later, all national member associations of FIFA were free to adopt this regulation.
Whether these were experiments in England and Scotland, I can only guess. However, I have no proof for this and I do not know whether the possible experiment was only carried out in England and Scotland.
After 1967 nothing changed for some years, although there were experiments and proposals for changes.
What the US-American football association USSF wanted to test in 1974 seems familiar to us: Five substitutions instead of – back then – two. It did not receive permission from The IFAB, but nevertheless the USSF carried out the test and was reprimanded for doing so. This association also tested – excessively long – games with three substitutions from 1978 to 1981. Again, The IFAB reprimanded the USSF because after three years there was still no report of the experiment. Apparently, the USSF found the two possible substitutions too few and increased the number behind the cover of experiments.
It was discussed at the 1986 Annual General Meeting that a maximum limit on substitutions should be removed in general. However, this discussion did not receive the necessary majority, nor did the 1988 proposal to allow the substitution of two field players and a goalkeeper.
The provision later known to us as the 2+1 rule was proposed again in 1993 and, after lengthy discussions, was postponed until the following year. Then, in 1994, this amaendment was accepted. And was replaced the following year by three substitutions independent of position.
The fourth substitution in the extra time
The idea was first submitted by the Scottish FA at an Annual General Meeting in 2009 but did not receive the necessary majority this year. It then took six years before it was proposed again. This time by FIFA, CONMEBOL and the national associations of USA and Sri Lanka. Although FIFA also submitted the proposal, the then FIFA President, Joseph Blatter, was very distanced from the proposal. He feared that such permission could open the Pandora’s Box. So, the proposal was first submitted to the newly established Advisory Panel of The IFAB, which has been independent for a year.
A year later, Blatter was no longer FIFA President, but Gianni Infantino, who, unlike his predecessor, was a great friend of the fourth substitution in extra time. The proposal was initially allowed as a two-year, worldwide experiment. The matches that went into extra time were analysed and a survey conducted. After one year the first reports were very positive, so that the minimum requirements and a Gold Standard for professional use were developed. The Gold Standard then became part of the Football Quality Programme.
Since 2018 there is officially the fourth substitution in extra time.
This was the last change concerning the number of substitutions until spring/summer 2020. Since then and until the end of July 2021, national associations are free to allow up to five substitutions at competitive matches. However, these must be made within three breaks (excluding the half-time break).
A special case: Substitutions in international friendly matches
For test matches, the number of substitutions since 1972 may be higher than for competitive matches. Initially, up to five substitutions were allowed. This number was increased to six for friendly matches in 2004. In 2005 it was added that more are possible in other matches, provided the teams agree on this before the start of the match. The referee must of course be informed in advance.
The Number of substitutes
But there are other areas of substitution where something has changed since the 1970s: The number of named substitutes has changed very little. In 1971 it was decided that the two substitutes had to be chosen from five previously named players. Today the competition rules specify how many substitutes may be named, with a minimum of three and a maximum of twelve.
The proposal to increase the number of the named substitutes from five to eleven was rejected in 1987 and 1990 and withdrawn in 1991. It was voiced by several national federations both before the 1986 World Cup in Mexico and the 1990 World Cup in Italy and said that a large number of more players could be allowed to play in a World Cup match.
In 1992, the proposal came back to FIFA, but with up to nine named substitute players. FIFA granted UEFA the right to test this variant, and to do so at the European Championship next summer.
In 2002 during the World Cup in South Korea and Japan there was the possibility of naming up to twelve substitutes as a trial.
Credits: Steindy (talk): Match of the Austrian Football Bundesliga – FC Admira Wacker Mödling (red shirts) vs. SK Rapid Wien (white shirts) 2-1 (0-1) on 2015-12-02 in Bundesstadion Sudstadt – The photo shows the forth official Walter Altmann (right) and Ilter Ayyildiz (left). https://commons.wikimedia.org/wiki/File:FC_Admira_Wacker_vs._SK_Rapid_Wien_2015-12-02_(177).jpg. CC-BY SA 3.0.
Further changes to the football substitution
After having been successfully tested in England and Scotland, return substitution in recreational and amateur football are possible since 2015.
The place of substitution
In 1972 it was decided that substitutions must always be made on the centre line. Before that, there were no more detailed rules on the place. And in 1973, it was added that the substitute may only enter the field after the substituted player has left it. Of course, all this only with the approval of the referee.
This remained unchanged for almost half a century, until the 2019/20 season. Now, the substituted player had to leave the field at the next point, in order to minimise the increasing time wasting caused by the substitution process. However, the substitute must still enter the field at the halfway line and only after the substituted player has left the field.
The most important data for the development of football substitution
- 19th century: There were no substitutions in the modern sense, but by mutual agreement, the possibility of replacing players who appeared too late with another.
- 1923: Substitutions are only possible in non-competition matches and in the event of a serious injury, and only after prior agreement between the two teams. The referee must also be informed of the agreement, as must the substitution during the match.
- 1930s: In continental Europe, it has become a sort of common law in some countries to allow substitutions beyond the 1923 regulation. The four British associations unofficially tolerate FIFA’s casual attitude.
- 1951-1954: There are experiments with substitutions of injured players at competitive matches. The experiments took place at a FIFA international youth tournament, the 1953 World Cup qualifiers and the 1954 World Cup.
- 1957: The IFAB allows national associations to allow substitutions of injured players at competitive matches.
- 1965: In England, up to two players per team can be substituted per match regardless of injury. From the 1966/67 season also in Scotland.
- 1967: The IFAB granted the national associations to allow the substitution of up to two players per team per match, regardless of injury.
- 1994: Introduction of the 2+1 rule: two field players and a goalkeeper may be substituted.
- 1995: Up to three replacements, independent of position, are allowed.
- 2015: The national associations may allow return substitution in recreational and amateur football.
- 2018: After two years as a global experiment, the fourth substitution during extra time is officially introduced.
- 2020: Until the end of December 2022, national associations may allow up to five substitutions from late spring, i.e. the restart after the Covid 19 season interruption. These must be made in a maximum of three interruptions of the game (excluding the half-time break)