LONDON, ENGLAND – AUGUST 11: Gold medallists Mexico celebrate during the medal ceremony for the Men’s Football Final between Brazil and Mexico on Day 15 of the London 2012 Olympic Games at Wembley Stadium on August 11, 2012 in London, England. (Photo by Michael Regan/Getty Images)
Mexico succeeded in the 2013 Under-17 World Cup. Will they do as well in Brazil in 2014? Not necessarily. Data on the last five World Cups show no clear relationship between the two series. To understand World Cup performance, a clearer connection comes from looking at FIFA rankings and the results of the Olympics.
Stories about the Mexican under-17 boys national soccer team have made everyone in Mexico proud – at least for a day. Three times in the past six years Mexico has gone all the way to the under-17 World Cup final, (Peru 2005, Mexico 2011 and United Arab Emirates in 2013). Mexico won gold in the Olympics (London 2012) and went to the semifinals in the under-20 FIFA Youth World Cup (Colombia 2011). So why did Mexico’s big boys’ squad struggle and scrape across the line to the Brazil World Cup in 2014?
To be fair, Mexico is hardly the only country in this situation. Since 1987, Nigeria has made it to six finals in World Cup tournaments for under-16/under-17 age level (that’s six out of 14!), two in the under-20s, and two more Olympic finals. At the World Cup, though, in 1994 and 1998, they barely made it out of the first round, and in 2002 and 2010 they didn’t survive the playoffs. In 2006, they didn’t make it to the World Cup at all.
Then on the other hand there’s Spain. The Spanish national boys teams reached the finals in 1991, 2003 and 2007 in the under-17 leagues, in both 1999 and 2003 with the under-20 national team, and they went on to the Olympic finals in 2000. Needless to say, Spain is the current World Cup champion.
Spain is a good place to follow some of the best players who are heroes of the Spanish national team. Andrés Iniesta, to take one, played with the under-20 national team in 2003. They took second place in the UAE World Cup of the same year. Xavi and Iker Casillas also played with the under-20 team, but in Nigeria in 1999. They both became champions. Bojan, who seemed to be the next big figure with Barcelona, made it through the finals in the 2007 under-17 World Cup.
These three cases obliged us to answer the question as to whether performance on the boy’s national teams (under-17, under-20 and under-23) later reflects on performance on the adult team. The answer is no, at least not by in terms of individual performance.
Analyzing the performance of all the national teams affiliated with FIFA prior to each World Cup, it seems that passing from non-classifying to a youth World Cup (under-17/under-20) and on to reaching the final (i.e. moving from worst to best performance) has no statistically significant effect on the probability of classifying or playing an extra match in the FIFA World Cup. That means that the performance of national teams by themselves doesn’t necessarily reflect on their performance in the World Cup. We studied the performance of the under-17 World Cup five years prior to the current World Cup, three years prior for the under-20 teams, and two years prior for the under-23 teams.
The performances of the under-23 (plus three overage players) national teams during the Olympic games before a World Cup do seem to have an effect, though a small one, on the FIFA national teams. Going from non-qualifying to the Olympics soccer tournament and then on to the finals, reduces the probability of a team not qualifying for the World Cup by 6.2 percent. It also increases the probability of reaching the World Cup final by some 0.08 percent. That means that a team’s Olympic games performance will be an indicator, though a small one, as to their performance in the FIFA World Cup.
Performance of the same national team in a past FIFA World Cup seems to have no effect on performance in the current edition. This could be due to the great volatility of some teams. For example, South Korea reached fourth place during the 2002 World Cup (in Korea and Japan). In 1998 they barely tied a game in the playoffs and in 2006 they never made it through the playoffs. A few teams, like Brazil, do show some fortitude over time. In 1994, 1998 and again in the 2002 World Cup, they made it straight through the finals. 2006 and 2010 saw Brazil climb through to the quarterfinals.
Despite all this, FIFA rankings in our model do seem to matter and often correctly predict World Cup performance. This could be because FIFA ranking considers all matches over the past four years, including friendly matches, qualifying matches, Confederation Cup matches and World Cup matches. These are weighted by the record of the opposing team from the time since the last match occurred and also by the records of the contending confederations.
The FIFA World Cup, according to our own analysis, seems a game only between the best-ranked nations. Changing a team’s ranking from 170 to the best ranked-team in the world, increases their chances of qualifying for the World Cup by a massive 82 percent, and by 5.5 percent of making it to the finals.
Given all that, remember that Mexico was ranked as the fourth best national team when the 2006 World Cup began. They lost to Argentina in the first round of playoffs. Nigeria, on the other hand, was ranked 11th best team in the world, and never qualified.
In reality, the only thing we can discern is that other circumstances are needed to catalyze the success of younger teams to later be reflected in their national teams. This could be due, perhaps, to the relative competitiveness of their domestic leagues, the personal development of players within their national clubs or the humility with which they accept whatever success they enjoy. Studying cases of such success, as in Spain, may very well be the key for countries like Mexico to finally boast a successful national team.
For now, the numbers say that soccer is a changing and sometimes even random game. Perhaps that’s exactly why we love it.
All data was obtained via Wikipedia and the official FIFA web page. For a comparison of teams, several assumptions were made. Results of Great Britain in the Olympics were identified with the results of England in the FIFA World Cups (rather than Scotland, Wales or Ireland). Also, results obtained by Yugoslavia, then Serbia and Montenegro were normalized and matched with results of current Serbia, rather than Montenegro. The under-16 World Cups of 1987 and 1989 were treated as under-17 World Cups. All Olympic country teams were treated as under-23 teams, despite that teams are allowed up to three older players. For this analysis, we used an ordinal logistic model with the dependent variable the number of games played in the World Cup, and the independent variable the number of games played in the five years prior to the under-17 World Cup, three years prior to the under-20 World Cup, and two years prior to the Olympic games and the FIFA ranking of May prior to each World Cup.