14/06/2018

Statistics indicate that Brazil and Germany are favorites to win the World Cup.

Getty Images

Adapted from a Phys.Org report.

Brazil and Germany are the favorites to win the 2018 World Cup. The two teams are almost on par in terms of chances of winning, according to a statistical model based on odds from betting houses.

The model was developed in Austria by professors Achim Zeileis, from the Department of Statistics at the University of Innsbruck, Christoph Leitner and Kurt Hornik, from the Vienna University of Economics and Business. Together, they have already successfully predicted the winners of several tournaments, including Spain as world champion in the 2010 World Cup and three of the four semi-finalists in the 2014 World Cup.

In the World Cup in Russia, Brazil leads with a 16.6% chance of winning the title. It is closely followed by Germany with 15.8%. "The most likely final, with a probability of 5.5%, is also a match between these two teams, giving Brazil a chance to make up for the dramatic semi-final of 2014," says Zeileis.

If the final is confirmed, it won't be easy for the Brazilian team. Statistics indicate that the probability of Brazil beating Germany is only 50.6%.

The researchers use the odds from 26 online betting providers, which, combined with complex statistical models, allow for a simulation of all possible variations and outcomes of the game, from the group stage to the final.

Besides the favorites, there are two teams with good chances: Spain (12.5%) and France (12.1%). The most likely pairings in the semi-final are Brazil vs. France (9.4%) and Germany vs. Spain (9.2%), with Brazil and Germany more likely to emerge victorious.

Russia, the host country, is seen by bookmakers as the 12th best team: the probability of Russia reaching the quarterfinals is 28.9%, but it is reduced to 2.1% to win the tournament.

"Naturally, bookmakers want to make money from their offers and therefore set their odds as realistically as possible, taking into account not only historical data but also the tournament draw and short-term events such as injured players," says Zeileis.

Before using the data for prediction, researchers adjust the odds set by bookmakers for profit margins (so-called "overrounds").

Betting odds give each team a basic probability of winning. From this, statisticians can determine the probability of a given team playing and winning against another. Combined with bookmakers' expectations, the comparison of odds of winning can be incorporated into a calculation model that can be used to simulate any possible game variant on a computer.

"Our model has the advantage of directly estimating the overall probability of victory for each team, while also implying 'survival' probabilities throughout the tournament. But we are far from a 100% accurate prediction," adds Zeileis.

The most likely prediction for EURO 2016, for example, was that the host country, France, would beat Germany in the semi-finals and then win the final. "If Gignac had scored the goal in extra time against Portugal, instead of hitting the post, our prediction would have been perfect," the expert assures.

But things turned out differently, and Portugal won in extra time. This shows that small things can often make the decisive difference in football. That's why predictions with high probabilities cannot be made.

"It is in the very nature of predictions that they can be wrong. Otherwise, football tournaments would be very boring. We only provide probabilities, not certainties," concludes the researcher.

The full forecast with interactive charts can be found here .