Since the final draw for the 2014 World Cup on 6 December, there has been discussion of which group is the toughest, the so-called group of death or, in German, the Todesgruppe. Of the eight groups, Groups B, D and G – with Spain, England and Germany, respectively – seem to be the main candidates. According to some of the German media, Germany’s group is the Todesgruppe, while for the English press, England are in the group of death.
One reason why England may be in the most difficult group is because it was a victim of the pre-draw in which one of the nine European countries in Pot 4 had to be moved to Pot 2, which initially only contained seven countries. In advance, many feared a horror scenario in which the European country transferred from Pot 4 to Pot 2 was a strong team and was combined with a strong group head from Pot 1 and another strong European country from Pot 4 in one group. This has actually happened, since Uruguay from seeded Pot 1, Italy from Pot 2 and England from Pot 4, along with Costa Rica, now comprise the strong Group D. However, despite England’s misfortune of being in the same group as the relocated country from Pot 4, Groups B and G are even stronger.
For the assignment of countries in the seeded pot, alongside the host Brazil, FIFA uses a flawed, wobbly list – the FIFA World Ranking Table. In this table from October 2014, Colombia (4), Belgium (5) and Switzerland (7) are ranked higher than England (11), the Netherlands (8), Italy (9) and even Brazil (10). In the sports sciences, the FIFA ranking has been criticised for its many revisions since its inception in 1992. Also, the FIFA ranking sometimes gives awkward results: Norway ranked second in 1993 and in 1995, the USA ranked fourth in 2006 and Israel just missed a top-ten ranking because of a draw against Latvia in 2008. A much better ranking table is available, the World Football Elo Ratings, which uses the same methodology as the Elo ratings in chess. In this contribution, we will use the Elo football ratings to interpret the final draw of the World Cup 2014.
The average Elo rating on the day of final draw (6 December 2013) of all 32 teams participating in the World Cup 2014 was 1,823; with Algeria having the lowest rating of 1,582 and Brazil the highest at 2,110. To give an indication of how differences in ratings translate into the probability of wining, if one team has a rating 100 points above another, it has a 64% probability of winning in a knock-out match between the two; a difference of 200 points corresponds to a win probability of 76% percent; and 400 points to a probability of 91%. England has the highest rating (1,906) in Group D, followed by 1,898 for Uruguay, 1,887 for Italy and 1,717 for Costa Rica. In the match between England and Costa Rica, the difference is 189 points in favour of England, resulting in a 75% probability of England winning (a tie can be considered as half a win, i.e. 70% percent win and 10% tie).
In a balanced division of countries into groups, the average rating per group would be close to the overall average rating of 1,823 for all teams. However, this is not the case. Group H, with Belgium, has the lowest group rating of 1,720, which is more than 100 points below the overall rating. The strongest group of all is Group B, headed by Spain, with has a group rating of 1,916, almost 100 points above the overall average. So there is a spread of almost 200 points in the average group rating between the strongest group B, headed by Spain, and the weakest Group H, headed by Belgium. England’s group ranks only third, because the average rating of of Group G (headed by Germany) is 1,877, 25 Elo points higher than that of Group D (with England).
One may argue that to proceed to the next knockout stage of the tournament, it is not so much the average rating of the group that is important, but only the strengths of the teams that may realistically be expected to compete for the first two places in each group. If we leave out the country with the lowest rating in each group, the England’s group still ranks third after Groups B and G. Another consideration is to anticipate which country is most likely to be the opponent in the first knockout stage, the Round of 16. For England this is most likely to be Greece or Colombia, while for the first and second teams in Group B it is likely to be Mexico and Brazil. From this perspective also, being placed in Group B must be considered more difficult than England’s group. To make a comparison with the seeding in tennis, where the top players are situated in such a way that they will only meet each other in the last stages of the tournament if they perform up to their expectations, at the 2014 World Cup some of the top teams will probably meet each other as early as the first knockout stage (such as Brazil and the Netherlands).
England’s group is special, however, since there is no other group in which the spread of points among the top three teams is so small. Due to the small differences in ratings between England, Uruguay and Italy, these matches are very balanced, with win probabilities for England of 51% and 53%. If the tournament evolves according to the Elo ratings, England will be first in their group, will beat Greece in the Round of 16 and will eventually be eliminated by Spain in the quarter-finals.