A system of machine learning (artificial intelligence), devised by Technical University engineers of Dortmund in Germany to predict the team that will win the World Cup, foretold that the German team is the favorite to win.
The system devised by Andreas Groll and his companions is based on a combination of automatic learning and conventional statistics, a method called “random forest” approach. In particular, they model the outcome of each match and use the results to deduce the most likely course the competition will follow.
This machine learning system (in Spanish, automated learning, one of the forms of artificial intelligence) results in some predictions that are somewhat different from those of bookmakers or other means. Thus, for example, the method used by Groll chooses Spain as the most probable winner, with a probability of 17.8 percent.
According to the study, if Germany exceeds the group stage, it is more likely to face strong opposition in the elimination phase. This is why the “random forest” method estimates that Germany’s chances of reaching the quarterfinals are 58%. On the contrary, Spain is unlikely to face strong opposition in the last 16 games, so it has a 73% chance of reaching the quarter-finals.
Thus, the Spanish and German teams are expected to play the World Cup finals. “Spain has slightly more advantage over Germany, mainly due to the fact that Germany has a comparatively high opportunity to leave in the sixteen round,” explains Groll.