What Ashley Madison Can Teach About Prediction Markets

By Ben Golden on August 31, 2015

(You know, besides how to cheat...)

The recent hack of Ashley Madison--a dating service marketed towards married people--revealed that almost no women were actively using the site. Rather, the site's almost entirely male userbase was paying to interact with non-existent, fake, or inactive female accounts. For some reason, this made me think about prediction markets.

A common challenge with prediction market projects--particularly those within an organization--is recruiting users, especially early on. Many users value the social element of prediction markets--seeing reactions to their forecasts--and this experience is hard to create when there are few users. Prediction markets have a network effect, meaning that as they grow, they're more valuable to each individual user.

One potential solution is to use bots--forecasting algorithms that look like real users but aren't. Introducing simple bots could potentially improve prediction market participation and accuracy, even if the bots themselves aren't particularly intelligent or accurate. New users would enter what looks like a highly active prediction market, even if there are very few other users. As the userbase grows, bots could be phased out. Within an enterprise, bots could be presented as a challenge--can you figure out which users are your colleagues and which are dart throwing monkeys? Can you forecast better than a mindless robot? Depending on company culture, this challenge could be seen as fun. (Or, possibly, offensive.)

For enterprise prediction markets, another potential solution is to introduce ringers--human users from outside the organization, who lack insight into the organization's operations (though they could have expertise in relevant topic areas). This again presents a challenge to employees, though this time they're trying to identify and outperform dart-throwing people.

Deceiving users isn't good practice, but if transparently adding bots or ringers to a prediction market could increase user engagement and forecast accuracy, it's an idea worth including in a prediction market administrator's toolkit.

If you liked this post, follow us on Twitter at @cultivatelabs.

Ben Golden/@BenGoldn is an Engineer at Cultivate Labs

prediction markets crowdsourced forecasting