How uncertain and full of incidents are FIFA World Cup games?

Dr Nicolas Scelles

Posted: June 16, 2014

In my last posting, I specified that my previous research showed that football games are the most uncertain with 85 to 90% of percentage of game-time with a difference between teams of no more than one goal[1]. Is this percentage true for the World Cup? I worked in the past about uncertainty of outcome but also incidents in the World Cup with a publication in French[2]. Why incidents? Because if a competition with only 0-0 has 100% of game-time with uncertainty, it is not sure that people really enjoy games! Why incidents and not goals? Because it is not sure that people really enjoy games with a lot of goals but only for the same team! Here are the main results I got. Since the 1990 World Cup, all the editions had a percentage between 85 and 90% except in 2002 (84.6%) with 1990 the most balanced (89.8%). Nevertheless, 1990 was also the edition with least incidents: only 1.35 per game whereas 1998 was the edition with most incidents (1.83 per game).

More interesting is to focus on “real” uncertainty and incidents. Consider North Korea against Ivory Coast during the 2010 World Cup: Ivory Coast won 3-0 with 1 incident (when it scored the first goal) and 20mn with uncertainty (until Ivory Coast scored the second goal). Actually, North Korea was already eliminated whereas Ivory Coast was never in contention for the qualification: before the game 2. Portugal 4 (+7) and 3. Ivory Coast 1 (-2), goal difference in all group matches is the first criterion after number of points to rank teams and 0-0 in Portugal-Brazil. Consequently, no uncertainty and no incident in North Korea-Ivory Coast!

By contrast, uncertainty and incidents are possible even with a difference of 2 or more goals. Consider Chile-Spain during the 2010 World Cup: after 37mn, Spain led 2-0. At this moment, the ranking in their group was as follows: 1. Spain 6 (+3), 2. Chile 6 (0, 2-2), 3. Switzerland 4 (0, 1-1), 4. Honduras 1 (-3). It means that with 1 goal for Switzerland, this latter would have become 2nd but with also 1 goal for Chile, Chile would have remained 2nd: 2. Chile 6 (+1, 3-2) and 3. Switzerland 6 (+1, 2-1). Consequently, Chile did not depend only of what could happen in Switzerland-Honduras but also of its own match even with a difference of 2 goals!

In taking into account “real” uncertainty and incidents, 2002 becomes more uncertain (87.5% of game-time with uncertainty instead of 84.6%) whereas 1994 and 1998 are less uncertain (respectively 82.8 and 80.9%). 1990 is still the most uncertain edition with 91.3% of game-time with possibility of incidents. In terms of actual incidents, 1994 is now better than 1998 (1.56 against 1.52) and 1990 is not the worst edition anymore (1.25 against 1.13 in 2006 and 1.19 in 2010).

What can we expect for the current World Cup? Uncertainty and incidents comparable to the last 2 editions (uncertainty between 85 and 90% and few incidents)? Games slightly less uncertain but more incidents like in 1994 and 1998? Or uncertain games and more incidents like in 1990 and 2002? The answer is still unknown. Nevertheless, even with between 80 and 85%, games would be quite uncertain. You understand why football games are so unpredictable and why my last words in my previous posting were: “I could not advise you to bet on the basis of my previous predictions!”



I[1] See for example: Scelles, N., Durand, C., Bah S. T., & Rioult, F. (2011). Intra-match competitive intensity in French football Ligue 1 and rugby Top 14. International Journal of Sport Management and Marketing, 9(3/4), 154-169.

II[2] Scelles, N., & Durand, C. (2010). Economic design : Optimiser l’intensité compétitive sous contraintes. Le cas de la Coupe du Monde de la FIFA [Economic design: Optimising competitive intensity with constraints. The FIFA World Cup case]. Jurisport, 99, 42-45.

About Dr Nicolas Scelles

Nicolas Scelles is lecturer at the School of Sport, Stirling University, Scotland. He holds a PhD in sports economics from the University of Caen Basse-Normandie, France. He has articles in international journals including Applied Economics, Economics Bulletin, International Journal of Sport Finance and International Journal of Sport Management and Marketing.