Nassim Nicholas Taleb may hate me for this post, but oh well, I’m going to go ahead with it anyway…
Prediction markets are a little-used, little understood tool used for – you guessed it – making predictions about things. At present, it is still on the fringes of use in business despite its growing body of evidence. James Surowiecki stoked the fires in his book The Wisdom of Crowds. The idea is that a “crowd,” or group of people, collectively know more than any one individual can know, no matter how intelligent that one individual is. By aggregating our knowledge, we can be better informed than even the most experienced, learned, intelligent expert.
Even the aforementioned Taleb, who would hate (and the guy has very strong opinions on these types of things) the idea of prediction markets (more on that in subsequent posts), does caution against ‘chasing the expert,’ so there may be some common ground there. Surowiecki and authors such as J.S. Armstrong point to the logical reasoning (with data to support) that suggests that averaging the opinions of experts is better than trying to choose the expert that is going to be right. The reasoning of these sound thinkers, combined with my own experience, is why I believe so greatly in the collaboration that exists among colleagues in workshop settings, or through the use of other collaboration tools.
The idea behind prediction markets is simple. By averaging the estimations (or guesses when I’m not writing academically) of individuals, we can achieve fascinating results. There is more to it than that, but I won’t bore all of you with the details. My own paper on the subject (see prediction markets link to the right), allowed me to predict, with 99.7% accuracy, admissions to a hospital for a given week. Jeff Severts has used these at BestBuy, and Google uses it to determine which websites provide the best content for your search criteria. I believe prediction markets have uses in the procurement sphere, which is the topic of an upcoming post.