How to predict a winner in 2018 FIFA World Cup

Remember Jabulani? Back in 2010‚ when South Africa hosted the Soccer World Cup‚ the octopus at the Two Oceans Aquarium in Cape Town was on target with his prediction that Spain would win.

The tournament also had Paul the octopus in Germany‚ letting his landsmann know how the team would fare throughout the tournament‚ and most of the time he was right.

For Russia 2018‚ scientists have gone all out to predict the winner‚ earmarking a Brazilian victory over Germany in the final. But their methods might boggle the layperson’s mind.

With Jabulani‚ it was simple and poetic. Jars holding the Dutch and Spanish flags were lowered into his tank just ahead of the final. He curled his suckered tentacles around “la Rojigualda” and the rest is history.

For Achim Zeileis at the University of Innsbruck‚ it is a matter of combining the odds from 26 betting websites with “complex statistical models”.

Then a simulation of “all possible game variants and results” is created.

That runs into millions. Because of all the rounds‚ statisticians had to “repeatedly play through the entire tournament millions of times – playing through every conceivable match pairing”.

After doing this‚ his team found Brazil had the highest chance of winning‚ at 16.6%‚ closely followed by defending world champions Germany on 15.8%.

“The most likely final is a match between these two teams‚ giving Brazil the chance to make up for the dramatic semi-final of 2014‚” he said.

South Africans aren’t convinced that their brothers from the global south will take top spot. Growth marketing company DCMN conducted a survey here and found Germany was earmarked as the winner by 26% of people – the highest for any country.

Over in Australia‚ a professor in decision-making and risk analysis has developed a Monte Carlo simulation of the competition‚ based on a technique developed by scientists in World War 2 who were working on the atomic bomb.

“The key idea is that rather than trying to work out every possible outcome of a complex system‚ enough possibilities are modelled to be able to estimate the chance of any particular outcome occurring‚” said Steve Begg at the University of Adelaide.

For Rose Hattingh‚ a South African now living abroad‚ a simple method not very different from Jabulani’s was her predictor of choice as a teenager.

“I made no predictions until the final‚” she said‚ “but then I would take out all my old dolls‚ bears‚ toys‚ you name it‚ and dress them in the colours of the two teams in the final.”

She would line them up against the wall‚ one team on the one side of her room‚ one team on the other.

“Then this was the best part. My brother and I would each throw a tennis ball hard so it hit the ground and then bounced around the room. We would know the winner from where the ball came to rest.”

And if it was a draw?

“We would keep throwing until we had a clear winner.”