Surprise -- there are limits to how well politics and economics can predict World Cup matches.
- By Daniel AltmanDaniel Altman is the owner of North Yard Analytics LLC, a sports data consulting firm, and an adjunct associate professor of economics at New York University’s Stern School of Business.
We gave it our best shot. Our Greatest Nation On Earth predictor for the World Cup in Brazil had some early successes, but the knockout rounds were rough sledding. A country’s ability to take care of its citizens and provide an attractive environment for investment may not be the ultimate determinant of its soccer prowess. But we’re not going to give up on the GNOE yet, since it may yet come up with the biggest winner of all.
One of the simplest ways to evaluate a predictor is to see how its pre-tournament rankings of teams squared with the final rankings in the World Cup. FIFA compiles these rankings after the tournament finishes, but they’ve already been decided for Brazil except for the top two spots. Essentially, teams are grouped by the round of the tournament when they were knocked out, and then by wins, draws, goal difference, and goals scored.
By this metric, the GNOE was only a tad better than a monkey throwing darts at a typewriter. The correlation between its pre-tournament rankings and the likely post-tournament rankings is likely to be between 0.1 and 0.2, compared with about 0.6 for FiveThirtyEight’s Soccer Power Index and close to 0.7 for FIFA’s own rankings. No one would have made money betting the GNOE.
Or would they? The GNOE was a reasonable predictor of success in the group stages, correctly picking 10 of the 16 teams that advanced. If we wiped the slate clean upon entering each subsequent round, how well would the GNOE have done in forecasting the outcome of individual matches?
In the 14 matches played so far in the knockout rounds, the GNOE would have picked eight winners correctly. That’s only a tad better than flipping a coin. As a betting strategy, with the same $1 stake on each of the 14 matches, the GNOE would have yielded a loss of $1.69, or 12 percent. But that’s partly because the two matches Costa Rica played, against Greece and the Netherlands, ended in penalty shootouts. Bookmakers classify those results as draws, so the GNOE would have received no credit for predicting the winners.
The GNOE can still redeem itself, however, with its forecast for the third-place match and the final this weekend. There’s a huge gap between the European and South American teams, so the GNOE is confident that Germany and the Netherlands will be victorious. If they do, current odds suggest the GNOE will once again be a net gainer, with a return of roughly $5.50 on two $1 bets easily covering its previous losses.
So what does all of this say about politics, economics, and soccer? On the face of it, not much. But I think this might change with time.
Great soccer talent can emerge from the dusty streets of far-flung slums, but developing that talent into players capable of winning the World Cup takes infrastructure, know-how, and money. It also requires legal, political, and economic institutions that ensure anyone with talent gets a chance to succeed.
Right now, some countries are gaining this capacity faster than others, and the GNOE picks up some of these differences. At the same time, the beautiful game is becoming more globalized. If soccer fever is uniform around the world, then the factors included in the GNOE will determine the top teams to an increasing extent. So watch out, Brazil, and enjoy it while you can, Argentina — this tournament may just have been a taste of what’s to come.