As we continue to use Huunu to work with Gannett on their prediction market game, The Ticker, continues to deliver incredible levels of predictive accuracy. Session 1 results from The Ticker were impressive to say the least, Ticker players were accurate 78% of the time across a variety of topics such as news, sports, and entertainment. When we isolated to sporting events, we found that the aggregated wisdom of the crowd accurately predicted results 81% of the time.
Even more impressively, the types of questions being asked in The Ticker varied. Players were asked to invest in point differentials as well as finishing order, and they did so with impressive accuracy as well. Notably, players predicted that the point spread of the USA Basketball team’s game versus Argentina would be 26.28 points; the actual point differential was 26 points. Similarly, when the USA played France, Ticker players predicted that the point differential would be 24.75 points; again, the actual point differential was 27 points. The market didn’t do as well when the USA obliterated Nigeria, but we’ll take the blame for that one. We didn’t actually give them the opportunity to guess correctly as we never foresaw that the actual point differential would go above 60 points. Not to worry, we don’t think anyone else saw that one coming either.
And Huunu and the crowd have done an amazing job predicting finishing orders as well. In a recent NASCAR race, Ticker players correctly guessed the finishing order of four drivers before the race had begun. These results are very reminiscent of the Heisman results that Huunu produced in December of 2011 when the Football Futures market predicted the correct finishing order of the players days before the final votes were tallied.
We aren’t the only ones have noticed the high accuracy levels that prediction markets can produce in the sports arena. Spann and Skiera devoted an entire paper to the idea in the Journal of Forecasting. You can find the paper here. (Martin Spann and Bernd Skiera (2009) Sports forecasting: a comparison of the forecast accuracy of prediction markets, betting odds and tipsters, Journal of Forecasting, 28(1), 55-72 (doi). The paper compares the accuracy of prediction markets, tipsters (experts), and betting odds both random and choosing home team advantage. Spann and Skiera conclude that prediction markets and betting odds (especially when choosing the home team) are best at predicting. Tipsters finish a distant third, proving what we knew from our years of experience, the crowd does know better.
Want to know more about Huunu for Research and Huunu for Media? Visit us on the web at www.consensuspoint.com