The FIFA football World Cup begins this Friday: but don’t worry soccer-haters, this isn’t an article about the game itself. Instead, we’ll be taking a look at attempts to predict the outcome using facts and figures.
Most people, let’s be honest, predict the results through a combination of guesswork and gut instinct. While there’s a theory many events on which people can gamble may be easier to predict by applying the “wisdom of crowds” philosophy to the cumulative wagers, that’s somewhat trickier for the World Cup where bets in individual countries are significantly distorted by patriotism outweighing judgment.
So what’s the objective way to make a prediction? Well, it might be quantitative analysis.
The term refers to a wide range of techniques used in the finance world which work by applying mathematical formulas and calculation to make predictions. Whether such techniques work depends on whether you believe human nature can be reduced to mathematics, particularly when aggregated on a large scale.
Several financial firms have chosen to test their skills by applying their techniques to forecasting the World Cup. The results may not guarantee we’ll know who will win, but the techniques and rationale are intriguing.
Danske Bank’s report is mainly about the effects the tournament will have on regional economies, but does predict the football. Its method works on a blend of football factors (historical performance, current form, home field advantage) and economic factors, the idea being that larger and richer populations are better placed to produce a skilled and well-trained team. The analysts looked at the last four tournaments to see how important these relative factors are. Their prediction: Brazil beats Germany in the final.
Goldman Sachs offers a less sophisticated method. It simply works on the basis of past performance and how that plays out with the groupings in this year’s events. It doesn’t forecast a winner but tips Argentina, Brazil, England and Spain as the semi-finalists.
JP Morgan bases its predictions on the three types of information it uses in stock market forecasts: valuation metrics; price trends; market/analyst sentiment; and company fundamentals. These translate as: the official FIFA rankings and odds from bookmakers; changes in both of these factors over the past year; performance in previous tournaments and recent games; and a combination of how widely odds on each team vary (the less variation, the more reliable the information is assumed to be) and the quality of recent opponents.
Applying these factors (in a ratio based on their perceived importance) shows Brazil as the strongest team. However, given that two close teams may well draw, the model also takes account of two factors which would play a role in a “tie-break” penalty shoot-out: goalscoring ability and penalty saves. Running the tournament match by match then sees Brazil eliminated in the quarter-finals on penalties, with England beating the Netherlands on penalties in the semi-finals and beating Spain on penalties in the final. Given that England have been eliminated on penalties in five of the last eight international tournaments they have competed in, that sounds dubious to say the least.
Finally UBS relies on three factors: performance in previous tournaments, home-field advantage, and performance in the past three months. It argues that socioeconomic factors such as population and wealth aren’t a guide to results. It doesn’t make predictions as such, but rather offers percentage likelihoods. One surprising figure is that it ranks South Africa as having the strongest likelihood of advancing to the second round, a prediction apparently based on the grounds that every previous host nation has done so: that may be an unreliable strategy given South Africa is also by far the weakest footballing nation to host the event. The firm predicts Brazil as most likely to win, with a 22% chance of success.
[Picture credit: Flickr user Coach_J]