Incredibly accurate Olympic medal predications
Posted August 13, 2012 12:43 pm.
This article is more than 5 years old.
COLORADO SPRINGS, CO (NEWS1130) – It’s better than looking into a crystal ball.
Pre-Olympic medal predictions from an economist seem pretty darn accurate, even if they don’t take into account any athletic data.
Colorado College’s Daniel Johnson did a pretty good job. His accuracy at the London Games was 97 per cent. He was a bit off for Canada, suggesting we’d bring home 17 medals, four of them gold, when in reality we were at 18 medals but just 1 gold.
Johnson was spot on, down to the medal for 38 per cent of countries. Nearly four of five were correct to within two medals, which is pretty impressive considering 962 were given out.
He under-predicted the number of medals for China, Great Britain, US, Iran, Jamaica and New Zealand. Japan also go more medals, however fewer gold then the prediction.
He had a perfect prediction of the rankings of the top five — US, China, Russia, Great Britain and Germany.
His model has shown uncanny accuracy. Since the Sydney games in 2000, his rate of success is 93 per cent. But surprisingly, he doesn’t analyze athletes or sporting news, rather it’s all about non-athletic data like per capita income, population and hosting advantage.
In previous years, the formula also included political structure and climate. Those were dropped for this year’s Games. Now there’s a “Cultural Specific Factor” that helps to correct the model’s historical under-predictions for nations like Australia and China.
You can take a look at Johnson’s current and past predictions, as well as Johnson and Ali’s published paper here.
