Spain will win the UEFA European Football Championship with a probability of 25.8 per cent, defeating Germany in the final. This is the forecast made by a team of scientists headed by Achim Zeileis of the University of Innsbruck. By applying a statistical model based on bookmaker odds, they also correctly predicted the EURO 2008 finalists and Spain as the 2010 FIFA World Champion.
From the official FIFA rating to astrological forecasts, in the run-up to the kickoff of EURO 2012 there are many predictions about who will win the European Football Championship. One statistical model seems to be more reliable when it comes to forecasting results: the bookmaker consensus model developed by Achim Zeileis, Professor of Statistics at the University of Innsbruck and his colleagues from the Vienna University of Economics and Business (WU Vienna). They have applied this model again in their most recent study on the UEFA EURO 2012.
The scientists obtained long-term winning odds from 23 online bookmakers, which in combination with complex statistical models were used for a simulation of all possible courses of the tournament and results. According to Zeileis and his colleagues, Spain has the highest probability to win the tournament (25.8 per cent), followed closely by Germany (22.2 per cent). In all likelihood, these two teams, which faced each other in the 2008 final, will also reach the final in 2012. Compared to past football championships, the scientists show a significant difference in this one: the winning odds of the other teams are considerably lower than those of the two favorites. For example, the Netherlands’ winning probability is 11.3 per cent and, is, therefore, the team most likely to rank third, followed by England, whose probability to win the UEFA EURO 2012 are only 8.0 per cent.
Bookmaker knowledge plus statistical expertise
Bookmakers base their odds on the most likely results. As experts they not only take historical data into account but also short-term events such as injuries. "As the odds are assigned in such a way that they predict the real results as accurately as possible on the one hand and ensure profit for the bookmakers on the other hand, they are an excellent basis for our study," says Achim Zeileis about the foundation of the model developed by himself and his colleagues Dr. Christoph Leitner and Prof. Kurt Hornik (both WU Vienna). "However, the quoted odds of the bookmakers have to be adjusted for stake and profit margins. After that we can deduct probabilities."
In the next step the scientists calculate or model the winning probabilities of the teams playing against each other. "The tournament schedule was already known at the time the bookmakers calculated their odds and, therefore, the winning probability of each team within its group has to be taken into account," says Zeileis. The scientists derive these winning probabilities from the classical statistics approach of pairwise comparisons. Together with the expectations of the bookmakers, pairwise winning probabilities are then added to the computer model, which then runs a simulation of every possible course of the tournament. "Compared to other models, ours has the advantage that we can not only simulate the winning probabilities but also the ’survival’probabilities of the teams," explains Zeileis. "However, we are still a long way from predicting the outcome with 100 per cent certainty," adds Zeileis. Thus, we can still look forward to exciting games until the final whistle blows in Kiev on July 1st.