Better FIFA ranking

From Master Projects
Jump to: navigation, search

About Better FIFA ranking

  • This project has been fulfilled.
  • This project fits in the following Bachelor programs: {{#arraymap:|, |xXx|bachelorproject within::xXx|,}}
  • This project fits in the following masterareas: {{#arraymap:Software Engineering, Technical Artificial Intelligence, Computational Intelligence and Selforganisation, Computer Science and Communication|, |xXx|project within::xXx|,}}


In 2010, there was a competition in which people came up with a ranking scheme for chess that is better than the currently used Elo ranking in predicting outcomes of chess matches. FIFA has a ranking scheme, too, for national football teams. Apparently, some countries exploit this way of calculation, and choose their friendly matches to climb higher in the rankings without risking too much, and that's how they can end up in better hats when qualification groups for events are selected. The task would be to create a new scoring scheme - based on historical data - that would better predict results of matches than the FIFA method. (Additional or alternative tasks can perhaps be to determine an optimal climbing strategy, or simulate how different behaviours would increase/decrease a team's ranking). Ideally, the student who takes it should have knowledge of data mining and machine learning, related subjects can also be useful (such as statistics, statistical data analysis, evolutionary computing, neural networks, etc.) Knowledge of academic writing is also desirable. If you have questions or ideas, feel free to contact me (Zoli).