AP file

When it came to scooping up medals at the 2010 Olympics, Canada led the pack in the gold category but Team USA won the most overall.

The gold medal for predicting the Olympic medal count goes to ... neither the math geeks nor the sports jocks. The clearest winners are the folks who shifted their forecasts after the Vancouver Games actually started and put their virtual money on Team USA. But the judges will have to make the final ruling - and that means you, gentle reader.

Seventeen days ago, when the Winter Olympics began, there was a clear split between economics-minded prognosticators who size up statistics to predict how different nations will divvy up the Olympic spoils, and sports-savvy observers who size up the actual athletes. Sports Illustrated, for example, predicted that Germany would take the most medals, while Colorado College economist Daniel Johnson's model showed Canada with the top tally.

That was the general pattern for pre-Olympic predictions: The economic models showed Canada on top, in part because those models had a "home-nation advantage" built in. The conventional wisdom in the sports world - including the conventional wisdom initially reflected in prediction markets such as Inkling and Hubdub - favored Germany.

In the end, neither country topped the medal count. Instead, the top slot went to the United States, with 37 total medals. Germany came in second with 30, and Canada was third with 26. If you consider only the gold medals, the order is reversed: Canada is on top with 14, followed by Germany with 10 and the Yanks with nine.

In his postmortem, The Wall Street Journal's Matthew Futterman said his economics-based forecast (which put Canada on top) "produced some strong results - and a few that were less than glowing." He made the numbers look better by lumping together the U.S and Canadian medals, and wondered whether the outcome pointed to "a new era of North American dominance in the Winter Olympics."

Thanks to Canada's gold-medal haul, a case could be made that the geeks were golden after all - at least from the Canadian perspective. The Calgary Sun praised the home team for "leading the world in wins" and hailed Johnson as a "sage" becaue his 27-medal prediction was so close to Canada's actual count.

But Johnson himself wasn't so quick to claim the podium's top spot: "It's pretty close, that's true, but I wish it were all strong news, and it isn't - for example, we were really off on Canadian gold, with the model predicting only five," he told the Sun.

Johnson said he might have to tweak the computerized model and add in more statistics, perhaps including nation-by-nation expenditure on athletics.

Another avenue for future research is to watch how prediction markets behave. Such markets allow "investors" to put play money down on various predicted outcomes - ranging from the winner of a presidential election to the virulence of the next flu outbreak. The market-based forecasts, which essentially rely on the wisdom of crowds, arguably yield more accurate predictions than opinion polls. 

At the beginning of the Games, the initial judgment of the jocks was reflected in prediction markets as well. Germany was favored as the top medal winner. But the beauty of the system is that the markets don't necessarily stand still. As more information becomes available, savvy investors can dump their stock in one proposition (Germany on top) and reinvest in a different one (U.S. on top).

AP file

Inkling's prediction market chart shows the price trend lines for the U.S. and German medal count crossing on Feb. 17. Volume is charted at the bottom.

That's exactly what happened during the Vancouver Games: Around Feb. 17, there was a dramatic shift in Inkling's medal market, with the price for Team USA's stock crossing above Team Germany's. A similar trend popped up on the Hubdub market. What's interesting is that the investment volume lagged behind the price shift, suggesting that the smart money switched to America's team first.

The factor behind the rise might have been the realization that U.S. skiers were doing better than expected. Or it might have been a more general realization that the U.S. team as a whole had the momentum in Vancouver. In any case, I would award the gold to the virtual investors in prediction markets, thanks to their late surge. The silver medal for Olympic predictions goes to the geeks, thanks to Canada's gold-medal showing. The bronze would go to the jocks who favored Germany.

Is that the way you see it? Feel free to weigh in with your rulings in the comment section below.