Since the 1998 World Cup Finals hosted by France, Goldman Sachs, that huge mega multinational investment banking firm, has used its statistical analysis skills—probably the same statistical analysis skills that it used to “mislead investors” during the 2007-2012 financial crisis that helped the company profit off of the mortgage market collapse—to attempt to predict the results of the globe’s quadrennial futbol spectacle, the FIFA World Cup.
With the exception of the 2002 tournament (and I’m just gonna say 1998 as well, because those predictions are nowhere to be found) the bank has successfully pegged two of the four teams to make it to the semifinals—it picked no correct semifinal team in 2002. The closest it’s come to picking the World Cup champion was in 2010, when the firm had Spain, the eventual champions, as runners up to Brazil. So it’s safe to say, as more and more data becomes available for them to plug into their model, the results are improving ever so slightly.
Photo source: Twitter
The methodology that Sachs uses is rather straightforward and probably not the best in the grand scheme of things. They essentially just take a team’s track record in all international games with the exception of “friendlies,” and uses their goals-per-game as the only real indicator. There are no controls for the squads’ current makeup (meaning any injuries going into the tournament, how individual players are performing, or even just the fact that Ronaldo or Messi are on a team), human element, etc. Here’s a more detailed explanation of how it works, via the document:
The predictions for each match are based on a regression analysis that uses the entire history of mandatory international football matches—i.e., no friendlies—since 1960. This gives us about 14,000 observations to estimate the coefficients of our model. The dependent variable in the regression analysis is the number of goals scored by each side in each match. Following the literature on modelling football matches, we assume that the number of goals scored by a particular side in a particular match follows a Poisson distribution.
I like this and don’t like it at the same time. What I like about it is that it is purely statistical, and it’s refreshing to get that kind of unbiased look at a tournament that breeds so much emotion and national pride. At the same time, though, there are so many other factors that come into play beyond how many goals a team can score, especially when we’re talking about teams travelling anywhere from a couple hundred to a couple thousand miles to play.
Using that wonky methodology, though, Goldman Sachs went ahead and predicted the results of each of the group stage matches, the subsequent results in the knockout round, and even listed the probability for each country to make it to different stages of the tournament all the way to the championship.
As it turns out, Brazil, the host nation for the 2014 World Cup, is the odds on favorite, according to Goldman Sachs’ model, and it’s not even close. The bank gave the hosts a 48.5 percent chance to win the whole thing, nearly doubling what odds-makers are offering.
In the group stage, the bank predicts the Auriverde will take care of both Mexico and Croatia by final scores of 4-1, and cruise past Cameroon 5-0. Their 13-2 goals-for advantage outpaces the next best squad by seven goals (Argentina has a 7-3 goals-for edge in the group stage). Here’s the complete list of group stage match results:
Of note in the chart above, England gets embarrassed in the group stage, bowing out of the World Cup without a win. Same for the U.S., but that wouldn’t be all that surprising given the fact they’re playing in this year’s Group of Death.
In the knockout round, Brazil again eases its way along, so take the time to look around the rest of the board:
What’s interesting here is how close they expect many of the other knockout games to be, notably Portugal and Italy drawing with Russia and Columbia respectively but moving on. Also, the all-South America final is something to dream about. One can only imagine what would be left of the countries after that match.
As for the economic impact of the World Cup, though, Goldman Sachs had this to say:
Looking at history, there is a clear pattern of outperformance by the winning team in the weeks after the World Cup final. On average, the victor outperforms the global market by 3.5 percent in the first month-–-a meaningful amount, although the outperformance fades significantly after three months. But sentiment can only take you so far, in markets at least—the winning nation doesn’t tend to hold on to its gains and, on average, sees its stock market underperform by around 4 percent on average over the year following the final. The message seems to be: enjoy the gains while they last.
Of note though, The only team to not follow the trend of outperforming the global markets following a World Cup win since 1972: Brazil, following their 2002 victory. Per Yahoo! Sports Canada, that team won when its economy was mired in recession and a currency crisis, and its stock market ultimately lagged the global market. And considering the current state of affairs in Brazil with the expenses for this World Cup and the 2016 Rio Olympics, lightning may end up striking twice if they do happen to come out on top next month.
The winners aren’t the only ones impacted financially by the result of the World Cup, according to Goldman Sachs. The runners-up and the host nation feel the effects of the tournament as well.
For the final losers:
Of course, the fans are always disappointed at losing at the last hurdle and, interestingly, the stock market doesn’t tend to react well either. In contrast to the initial post-match rally that the winners tend to enjoy, the runners-up seem to experience a post-final bout of the blues. The average relative outperformance of the runner-up is 2.0 percent over the first month. However, the average here is misleading as it is heavily skewed by Argentina in 1990, which enjoyed a 33 percent outperformance in the month after the final (that said, this bounce should be considered in context, since it followed a collapse in its stock market and currency that had reduced the value of Argentine equities by over 90 percent in US$ terms between October 1988 and early 1990). Aside from this epsode, 7 of the 9 other runners-up underperformed over the first month with an average underperformance of 1.4 percent. Interestingly, the poor performance doesn’t stop there. Most of the World Cup runners-up have seen their stock markets continue to underperform, with an average relative fall of 5.6% over the first three months.
And the host nation:
The focus on the host tends to bring with it both pride and confidence. As with the winner, there also tends to be a positive impact that lasts for a few weeks. In every case, the host country has enjoyed an outperformance of its stock market in the month after the event. The average outperformance is 2.7 percent, although this tends to fade fairly quickly, with nearly half underperforming over three months.
Being the host and the winner is the ultimate goal, but that’s only happened in six of the 19 World Cup tournaments played. While the early data shows those countries experienced better than normal stock performance in the first month, there isn’t enough data to be able to draw a real consistent conclusion about the ongoing benefits there.
All of Goldman Sachs’ predictions and models will be put to the test soon enough. The first match of the 2014 FIFA World Cup will kickoff on Thursday, June 12th at 2pm.