« Back to the Future in the soccer World Cup: Chile wins 1-0 or 3-1? | Main | Will June 27 become ‘Instant Replay in Soccer’ Day? »
Scrap FIFA World Soccer Ranking: Geography and Governance predict World Cup results
By Kaufmann | June 25, 2010 1 Comment »
In its own World Soccer Federation portal, FIFA.com, boasts: ‘since 1993, the FIFA/Coca-Cola World Ranking has become a regular part of international sports and an important indicator to find where teams stand in world’s football’s pecking order’…
Well, not quite, as it turns out, if judging by the results from an analysis of the Group competition stage that has just concluded in the football World Cup currently taking place in beautiful South Africa. A total of 32 teams qualified for the World Cup. They were divided into 8 groups of 4 countries each, competed against each other, playing 3 games each, for a maximum of 9 points. The top 2 teams in each group are now advancing to the next stage of 16. Eight games will take place over the next 4 days, starting on Saturday with Uruguay playing South Korea, and then later in the day Ghana plays against the US. And on Sunday Germany plays England in the earlier match, and so on until this coming Tuesday. The winner in each one of these 8 games advances to the Cup’s quarter finals, and so on.
Unless the group stage that has just concluded turns out to be a major anomaly, the instant analysis we have made of the performance of these 32 teams in this Group stage casts serious doubt on the FIFA rankings of teams. FIFA ranks over 200 country teams; its latest ranking was in late May shortly before the World Cup.
In these FIFA rankings at the outset of this World Cup, Brazil were naturally at the top, followed closely by Spain in 2nd place. Italy was 5th, France was 9th, Russia 11th, Egypt 12th, Greece 13th, USA 14th, Serbia 15th. Of these, neither Russia nor Egypt even made it through the preliminary qualifiers to the World Cup (so they were absent from the Cup’s Group stage). Italy, automatically qualified for being the defending World Champion and (kind of) played in the Group stage. But not only it failed to advance beyond the Cup’s Group stage; it finished last in its group. For good measure, highly ranked (by FIFA, that is) France failed miserably as well, and Serbia did not advance either.
In sharp contrast, countries ranked lower by FIFA, like Chile (18th), Paraguay (ranked 31st!), Ghana (32nd!), Slovakia (34th!), Japan (45th!!) and South Korea (47th!!), are all advancing to the next stage of the World Cup, having been successful in the Group stage of this football competition.
So if you were making bets on performance so far on the basis of FIFA rankings, you would not be faring well. In fact, some interesting results emerge from a simple and instant statistical analysis just done based on the number of points accumulated by each team in the Group stage of the competition.
1. FIFA’s ranking prior to the World Cup has little power in explaining the number of points accumulated by each one of the 32 teams; in fact FIFA’s rankings only explain about 15% of the variation in total points across all teams.
2. If you were betting on two teams which were 12 positions apart in the FIFA ranking prior to the Cup’s start (say the difference between the 13th and the 25th in FIFA ranking), the number of additional points accumulated by the team ranked much higher would have amounted to less than half (on average).
3. In sharp contrast, the region of origin of the country seems to do much better (twice as well as the FIFA ranking) in explaining how many points each team accumulated. FIFA ranking totally over-estimates the disappointing performance of Africa (by over 1.5 points!), and of the big European soccer powers (almost 1 point).
4. Conversely, FIFA rankings would have seriously under-estimated the number of points that surprising Asia would have accumulated in its 3 games (by almost a point, and even more so we focused on ‘globalized’ Asia, e.g. without North Korea). To an extent, FIFA rankings also under-estimated the US performance so far (by half a point).
5. By far the biggest outlier is another region altogether, however: South America, which FIFA’s rankings would have predicted that they would have accumulated a middling 4.5 points on average. Instead, all 5 South American teams (Brazil, Argentina, Chile, Paraguay and Uruguay) have succeeded in advancing beyond the group stage, accumulating an average per team of almost 7 points. FIFA rankings would have resulted in under-estimating each South American country’s points total by almost 2.5 points!
6. Contrary to convention, a very large economy (with very large financial and human resources) does not seem to explain how countries have fared in the group stage so far.
7. Instead, the quality of governance of a country (rule of law, corruption control, voice and accountability, etc., for which we use the WGI) does explain performance in the group stage. A country with relatively good governance could have accumulated almost 2 points more than a country with relatively poor quality of governance (difference of 2 standard deviations).
From this perspective, perhaps then it is not such a surprise that North Korea, Italy, Greece, France, Nigeria, Cameroon, Honduras and Cote d’Ivoire are going home early. [Of course this is not airtight: the well governed Swiss and Danes are also going home, for instance, perhaps victims of some European contagion...].
Obviously much more data and analysis is needed to ascertain robustness of these results and to see whether they are likely to apply more generally. And a country highly ranked by FIFA (or by anybody else for that matter) may well be the eventual winner of the World Cup, but that is no great feat. At a practical level the World Cup results so far raise serious questions about the usefulness of the existing FIFA rankings. A revamp is called for.
And at the metaphorical level, it suggests that this football game is pregnant with much larger messages. Sheer size or economic might in itself does not assure that the country is highly successful in soccer.
Governance matters, and below the lofty national level as well: some teams, such as Ghana, Chile, South Korea, the Netherlands, Japan and the US, thanks to good governance and leadership, have managed to subscribe to the motto that ‘the whole can be much more than the sum of its parts’, contrasting teams like Italy, France and Cameroon, where they have shown how much less the whole can be than the sum of its parts when misgovernance is rife.
South American ascent, European descent, African disappointment, unexpected Asian surprise. A broader metaphor as well?
Update, Saturday June 26th, the day the knockout stage started. Uruguay beat South Korea 2-1, continuing the Latin American ascent and stellar performance so far from all 5 South American teams (from COMNEBOL), while FIFA’s 47th-ranked South Korea represented Asia rather well in this match among the remaining 16 teams. Then Ghana, ranked 32nd by FIFA, won 2-1 against the United States (ranked 14th). Ghana was the only African country that had made it pass the World Cup’s Group stage to this early knockout stage. It now proudly moves on to the quarter finals, when it will play Uruguay (ranked 16th by FIFA). The challenge to FIFA’s world soccer rankings continues.
Update 2, Sunday June 27th, the second day of the knockout stage. A day that may become history in soccer, and finally force FIFA to reform and jump aboard the 21st century. What takes place on Sunday is so important that it deserves its own blog entry (here).
Topics: Corruption, Measurement Frontiers, Rule of Law | | 1 Comment
June 28th, 2010 at 5:46 am
Brilliant! It’s there for all to see. Clear lines of accountability, meritocratic system of talent management, and to excessive dependence on the left- or right-wings.
As an Englishman, is some consolation that my country is better governed off the pitch, than on it…