South African-based data analytics company, Principa, is applying predictive analytics and machine learning to predict the winners of the ongoing Rugby World Cup.
The Rugby World Cup tournament is taking place in England and SA's next game is on Saturday, when the Springboks take on Scotland.
"Although our models retrain every day and the margin may change, the current prediction is a South African win by 11 points. The bookies currently have the margin at 14 points in favour of South Africa," says Jaco Rossouw, director of delivery at Principa.
The company has signed up two internal teams of data scientists onto sports prediction site SuperBru.com, as an exercise to put theory into play in this year's Rugby World Cup, Rossouw adds.
By applying the same principles used to predict customer behaviour for the company's financial services and retail clients, the two teams of data scientists are vying against each other to develop algorithms and predictive models that can predict the outcome of the matches with the highest accuracy.
"Our teams of data scientists have looked at data from over 6 000 matches played by 99 teams going back 20 years to identify patterns that are highly probable to repeat in future," Rossouw says.
Some of the statistics and patterns the data scientists have discovered are New Zealand has won 84% of the matches the team played compared to 66% wins by Australia; 62% wins by SA, and 61% by England.
The data scientists say New Zealand scores an average of 36.9 points in a match compared to 28.9 by Australia, 28.5 by England and 28.4 by SA; and on average about 20% of all matches were upsets - where a higher ranking team was defeated by a lower ranking team.
"We've done very well in a few games, notably England vs Fiji (correct outcome); France vs Italy (out by two points); England vs Wales, we predicted the upset and were out by one point," says Rossouw.
"In almost all the other games, we did better than the bookies. The modelling process is complex, but one of our initial models indicated that a Japan upset in the SA vs Japan match was quite possible."
According to Rossouw, Principa uses the machine learning modelling process. Essentially, these are predictive models that are regularly retrained and adjusted, he explains.
He points out that although there are a great many opportunities to use this technology, the company will look at using it in the direct and digital marketing space and contact centre sector.
Share