The FIFA World Cup 2018 draws took place at Kremlin, Russia on December 1, 2017. There are 32 teams in the World Cup broken down into 8 groups of 4 teams each. The UEFA, European football’s governing council has 14 teams qualified. CONMEBOL (South America), CAF (Africa), and AFC (Asia) have five participating teams each while CONCACAF (Central and North America) has 3 teams qualified.
All teams are split into four different pots based on their FIFA world rankings, with Pot 1 being the strongest while Pot 4 being weakest. The host nation, i.e. Russia this year, is automatically placed in Pot 1.
FIFA has slightly altered the draw system this time round, which will prevent any “Group of Death” kind scenario in the group stages. Earlier, the pots were determined on a geographic criteria, however now, they are seeded solely based on their rankings.
Below are the seeds for all 32 teams:
The groups are then decided on the basis of a draw from each of these pots. Russia, being the host country was seeded into Pot 1, and hence was impossible to be grouped along with a powerhouse like Brazil or Germany. Based on ELO ratings – a measure of team’s quality that takes into account factors such as margin of victory, game importance, game location amongst several others, Russia’s group – Russia, Uruguay, Egypt and Saudi Arabia has an average rating of 1720, which is 98 points worse than the average of all World Cup teams. Additionally, this is the largest difference in group strength and WC average ever seen for any group in the World Cup since the previous decade.
Russia’s luck shone each time another group member was added from the pots. After Uruguay, Egypt, second weakest team of Pot 3, and Saudi Arabia, the weakest team of the WC based on both ELO and FIFA rankings were pooled into Group A, making it by far the most easiest of the Groups in FIFA WC ’18.
Using a combination of ESPN’s SPI (Soccer Power Index) and ELO ratings, we predicted the chances of each team advancing into the knockout stage. In SPI, each team is assigned an offensive and a defensive SPI rating, which is expressed in terms of number of goals it is expected to score and the number of goals it is expected to concede. Thus, a higher offensive SPI rating is good, while a higher defensive SPI rating is bad. After weighing the SPI and ELO ratings, we then ran Monte-Carlo simulations to play out each group fixture 10,000 times for each individual match forecast. For the group predictions, the Monte Carlo simulations remained static.
Following are the chances of each of the teams advancing into the knockout stages: