AAs the 2022 World Cup in Qatar begins to take shape – including the qualifying this week of reigning world champions France – more and more football fans are starting to look beyond and worry rightly so, the new format chosen by FIFA for the 2026 edition, which has many flaws. Fortunately, there is enough time for FIFA to adopt another one.
If the World Cup in Qatar will be played on the classic format of 32 teams, divided into eight groups of four, the 2026 World Cup, organized in the United States, Mexico and Canada will be very different: there will be 48 teams and the FIFA announced in January 2017 that they would be split into sixteen groups of three. In each group, the two best teams will advance to the round of 16.
However, this format has many flaws. In particular, it increases the risk of collusion: the two teams playing last in their group could have an interest in arranging, explicitly or tacitly, in order to obtain a result which qualifies them both, at the expense of the third team in the group. , who will have already played his two group matches.
There is a great risk that many teams will refuse to attack each other. Remember FRG-Austria at the 1982 World Cup [les deux équipes s’étaient arrangées pour éliminer l’Algérie]. FIFA, aware of the problem, considered eliminating draws in order to reduce the risk of collusion, but my calculations showed that even then the risk would remain high.
This format is also wobbly and sub-optimal since the group stage would require three days of competition, but on each of these days one in three teams would not play. Not only that “Would break the rhythm” competition; it would also reduce the sporting spectacle, as well as the economic value of the World Cup.
Televisions and sponsors would pay more if all teams played each day of the group stage, and the extra matches would generate more ticket revenue. Also, in this format, sixteen teams would be eliminated after just two games, compared to three in the current format.
Eight groups of six, divided into subgroups of three
To safeguard this formidable sporting spectacle that is a World Cup, FIFA must therefore adopt a better format. In 2019 I had suggested several options. I present here the format which seems to me the best.
I propose to divide the 48 teams into eight groups of six, but so that the tournament can still hold in a month, instead of facing the other five teams in the group, each team would only meet three.
To do this, each group of six would be divided into two sub-groups of three, as balanced as possible, competing in cross-competition. For example, group A would be divided into two subgroups A (a) and A (b), and each team from A (a) would only meet the 3 teams from A (b), and vice versa.
The top four teams from each group would qualify for the Round of 16; three of them could come from the same subgroup. Qualifying the top two teams from each subgroup would be potentially unfair, for example if all three teams from the same subgroup win all of their matches.
For the ranking of a group to be meaningful, it is important that the two subgroups are balanced. To do this a six-hat system of eight teams could be used in the toss.
For example, subgroup (a) would be made up of teams from hat 1 (the top eight in FIFA rankings), hat 4, and hat 5 or 6 (the weakest eight), and subgroup (b ) of teams from hats 2, 3 and 5 or 6. To best balance the two sub-groups, one could choose to draw the third team from the two sub-groups in hat 5 or hat 6 depending on the sum FIFA ranking points of the two teams already drawn in each sub-group.
In each group, the three matches of Day 3 would be played simultaneously, reducing the risk of collusion. On each day of the group stage, all teams would play. Each team would be guaranteed to play a minimum of three games at regular intervals.
Fairer, more efficient and better paced
This format would be fairer, more efficient and better paced. It would also increase the sporting interest of the group stage and financial income for FIFA and its member associations, offering more matches (104, against 80 in the sixteen groups of three formula) without lengthening the competition.
Another advantage: the final table would be simple, symmetrical, readable: the group winners would meet group fourth (for example A1-B4), and the second group of group third (for example, A2-B3).
Another format would be to divide the 48 teams into twelve groups of four. The best two from each group and eight best third in a group would advance to the Round of 16 (104 games in total); or else the eight best group winners would qualify directly for the Round of 16, while the other four group winners and the twelve runners-up would face off in the Round of 16 (96 games in total).
But in these two formats, since the number of groups (twelve) is not a power of 2, the final table would be wobbly and subject to arbitrary choices: in the first case certain group winners, chosen arbitrarily, would face third group in the round of 16, while others would play against group runners-up.
The eight-group format of six is fairer, less arbitrary and more readable. It would solve the problems posed by the sixteen group of three format. The World Cup could still be held over a month, while continuing to guarantee each team at least three days of rest between two matches. Six matches per day would be staged during the group stage, which would require eighteen stadiums (compared to sixteen currently planned).
It’s not too late for FIFA to adopt a better format for the 2026 World Cup.
Julien Guyon is a mathematician and football fan. A quantitative analyst, he is also an associate professor in the mathematics department at Columbia University and at the Courant Institute of Mathematical Sciences at New York University. His work is available on his web page and on his Twitter account @ julienguyon1977
