Microsoft’s “artificial intelligence” defeated Google in terms of accuracy of predicted scores in the World football cup 2014. So, Microsoft’s oracle operating on the Bing search engine successfully predicted the outcome of 15 out of 16 games of the elimination stage of the FIFA World Cup.
Microsoft’s rival Google fell behind in the prediction stakes. Google’s oracle fell foul of the Germans when it predicted France to beat them in the previous round. However it did accurately predict the outcomes of both the semifinals.
China’s Baidu states that like Microsoft it has also proved to be quite accurate with 15 out of 16 predictions setting right. However, Baidu’s forecast was made public post factum last week.
Baidu insists that it surpassed Microsoft in terms of accuracy of all matches’ scores of the tournament including the group stage. So, Baidu made 58.3% right predictions of all matches against 56.3% forecasted by Microsoft.
Google launched its oracle called the Cloud Platform in late June after the group stage. Microsoft, Google, and Baidu used their own methods of predictions, but all of them are based on the Big Data concept.
Google explained the methodology used for predictions, “Using data from Opta covering multiple seasons of professional soccer leagues as well as the group stage of the World Cup, we were able to examine how activity in previous games predicted performance in subsequent ones. We combined this modeling with a power ranking of relative team strength developed by one of our engineers, as well as a metric to stand in for hometeam advantage based on fan enthusiasm and the number of fans who had traveled to Brazil. We used a whole bunch of Google Cloud Platform products to build this model, including Google Cloud Dataflow to import all the data and Google BigQuery to analyze it.” Microsoft’s prediction model is based on an array of variables such as previous win/loss record, the strengths and weaknesses of teams, results in the qualifying stages of the competition, and the margin of victory. Besides, it takes into account how close a tournament to a team’s home country is as well as the type of grass on a field and weather conditions on the day of a match. In addition, Microsoft considers data which reflect wisdom of crowds.
Microsoft’s oracle is not only restricted to the World Cup or any other sporting event. Earlier, it was tested to predict the outcomes of reality shows like American Idol and The Voice. Winners of these shows are determined by TV audience voting. To predict voting, Microsoft monitored the audience sentiment in social networks. After successful experience with predicting the World Cup results, Microsoft is going to continue its experiment with forecasting the U.S. Congress election this autumn, in particular.
As for Baidu method, it monitored 37,000 matches played by 987 teams for the recent 5 years, China Daily reports. Apart from the traditional data (such as relative team proficiency, latest scores, and statistics of world cups) Baidu used bookmakers’ quotes.
China Daily affirms that football predictions made by huge hi-tech companies turned out to be more reliable than the econometric forecast from investment banks. Thus, Goldman Sachs, Danske Bank, PricewaterhouseCoopers, as well as several other finance companies posted their predictions before the World Cup. However, they were far from the truth. In particular, the econometric forecast was wrong predicting Brazil’s victory at the World Cup held at home. Besides, it was not able to foresee early elimination of the Spanish team being the former world champion.
