Can anyone accurately forecast the result of the London Olympics while they are a week away?
Even more daring, can they forecast results without in depth knowledge of athlete training?
Economists at Ruhr-Universität Bochum are taking a shot at it and, if it sounds like complete madness, we'll go ahead and spoil it and tell you they believe that China, USA, and Russia will top the score table with 102, 100, and 71 medals, respectively, so you can move on to another article.
But if you are interested in the nuts and bolts, read on. You can decide if this is mumbo-jumbo or solid science.
The calculations of the Bochum economists give them confidence that forecasts are possible by applying econometric models. They have entered political, economic, demographic and cultural data into their model and have retrospectively ‘forecast’ the medals table of previous games. The forecast is amazing: It correlates up to 97.4 % with the actual medals table of participating countries of the Athens games in 2004 and up to 96.9 % of the Beijing games in 2008.
Downside to that method: programmers often build models that solve test problems and they fix the models until it matches the measured data. If predicting the past were all it takes to hedge against the future we would not have had financial panics once per decade in the last 100 years.
Interestingly, they believe spending money can buy medal success. Can it? Losers in elections always blame money. They may have a point. In 2008, when American President Barack Obama was allowed to raise and spend twice as much money as his opponent Senator John McCain, academics phrased it as "charismatic candidates, motivated donors, and creative uses of the Internet" but that all takes money and Obama was not crippled by McCain's ethical stance on campaign finance reform.
Before the 2008 Games in Beijing, China dedicated more than US $4.5 billion to replace the USA as a ‘sports superpower’ and the British government also considerably increased spending in anticipation of the 2012 Olympics. Yet money alone cannot do it. As any small school can tell you, population counts in increasing the chances of having a gifted athlete too. Populous countries have a larger number of talented and therefore successful athletes than small countries; wealthy countries can spend more money on their athletes than countries with comparatively low per capita income and there are outliers; countries which would not have expected great success, such as Jamaica in 2008 in Beijing.
Finally, being economists, they invariably found that countries with socialist systems or a socialist past were more successful in collecting medals. That explains the recent success of China and the past success of the USSR but not anything at all about the US.
Women from countries with emancipated societies and equal opportunities in the workforce are more successful in the Olympics than those who come from societies with a patriarchal character - though if anyone knows of a patriarchal society that has the population and the wealth to be competitive, they know of it in some other universe. In their model, they factored in the labor force participation rate, the fertility rate, the number of years since women had the right to vote - basically, countries where Islam is the main religion aren't going to do well.
Is there a home field advantage?
South Korea was an example of a home field advantage really mattered - mostly because it involved a level of cheating by local judges on an unprecedented scale. Outside cheating, it is real?
The respective host country has increased chances of success, they found - and so do athletes of the future host country, primarily from the early expansion of athletic support in their home country when preparing for their role as host. Climatic conditions of the home country also have an effect on the athletes’ medals success. Because of their poor training conditions, athletes from countries with an extreme climate are disadvantaged compared to participants from moderate climate zones, especially with respect to outdoor sports. Why, then, do Africans win so many long distance races? It's a mystery of economics.
Given a home field advantage (spending), they expect that the British will take home more medals than before (57) and Brazil (28) will increase its medal standing because the games will be held in Rio de Janeiro in 2016. The German team will win 12 % fewer medals and slide down to 36 medals in the medals table.
Based on the results of the empirical analysis, a forecast for the number of medals potentially won by the participating nations in the 2012 Olympics can be predicted. The top 15 nations and the forecasted number of medals: 1. China (102 medals), 2. United States (100), 3. Russia (71), 4. UK (57), 5. Australia (43), 6. France (39), 7. Germany (36), 8. South Korea (31), 9. Cuba (29), 10. Brazil (28), 11. Ukraine (28), 12. Italy (27), 13. Japan (27), 14. Belarus (19), 15. Spain (19).
Don't bet the rent money.
The detailed forecasts have been published at: https://sites.google.com/site/olympicmedalspredictions
- PHYSICAL SCIENCES
- EARTH SCIENCES
- LIFE SCIENCES
- SOCIAL SCIENCES
Subscribe to the newsletter
Stay in touch with the scientific world!
Know Science And Want To Write?
- Family Holiday Survival: 12 Ways To Deal With That Climate Change Denier
- A Vegetarian Carnivorous Plant...Wait, What?
- Wave-Particle Duality And Quantum Uncertainty - Two Sides Of The Same Mystery?
- Guest Post: Ben Allanach, On Open Access
- Falsifiability And The Integrity Of Physics
- Enlightenment Finding Eternity In The Now
- How Does Prostate Cancer Happen?
- "Isn't it all chemically laden? What food do you eat that has no chemicals in it? I suppose..."
- "So, Mr. Martinez, it seems that since medicines actually have some hidden risks even after testing..."
- "1) Give them more booze. 2) If they still can talk, give them more booze. 3) If they are still..."
- "Some cultures delight in eating rabbit. In much of Europe and parts of North America the meat sells..."
- "Look, I don't understand the whole Nobel prize thing, but I can only present the data I know.  ..."
- One in every three people with type 1 diabetes produces insulin years post-diagnosis
- Ninety-eight new beetle species discovered in Indonesia
- Journal of the American College of Cardiology notes most accessed papers
- Researchers discover new genetic anomalies in lung cancer
- Radiologist recommendations for chest CT have high clinical yield