The winner of the World Cup 2018 that starts today is one of the things everybody will talk about in the next month. Due to the interest shown so far, the final tournament in Russia is likely to be the most widely viewed sporting event in history, more popular than the Olympics.

The best way to find out what team may be the winner is by taking into account bookmaker’s odds. They are using professional statisticians to analyze the probability of any outcome in every single match. Combining the odds that different bookmakers give is an even better estimate. This has given as clear favorite Brazil, with 16.6 percent chances to win, followed by Germany 12.8 percent and Spain with 12.5 percent.

In recent years researchers have developed machine-learning techniques that can outperform conventional statistical approach. Andreas Groll from the Technical University of Dortmund in Germany and some of his colleagues used a combination of machine learning and conventional statistics in order to identify the winner, a method called random-forest technique.

The technique is based on the idea that future events can be determined by a decision tree in which an outcome is calculated at each branch by reference to a set of training data. The random-forest is using the outcome of random branches, not every single one, and it does that many times over, each time with a different set of randomly selected branches.

In the case of the World Cup, the researchers simulated the tournament 100,000 times. For a start, the winner came out to be Spain with a probability of 17.8 percent. However, due to the competition met in the early stages, Germany was seen to face strong opposition in the 16-team knockout phase and this is why Germany’s chances to reach the quarter-finals are at 58 percent. Spain, who will not face strong opposition in the final 16 has a chance of 73 percent to reach the quarter-finals. If both pass through the final 16, Spain has only a small chance to win the Cup.

And according to the 100,000 simulations and to the most probable tournament course, Germany will win the Cup. Basically, at the beginning of the tournament, Spain has the best chance to win, but if Germany makes it to the quarter-finals it becomes the front-runner.

The first match today, that we can see on TVR, is between Russia and Saudi Arabia. None of them looks like to make it to the quarter-finals.

The post The winner of the World Cup is to be decided between Spain and Germany as calculated by machine-learning appeared first on Business Review.