• by Michael Cornfield , Op-Ed Contributor, December 16, 2014

Pivit succeeds InTrade, the Ireland-based company that enabled web users to buy and sell bets on election outcomes. InTrade processed more than $200 million in 2012 U.S. election futures. Scandals capsized the site the following year, a fate Pivit intends to avoid by not accepting money. Instead, it seeks those willing to “battle your friends” during live events for prizes and bragging rights. Users see a percentage for a likely result of an election, sporting match, or other media happening and can click “up” or “down” from one to ten times, depending on their sense of how accurate the algorithmically determined probability will prove to be.

Pivit says it updates the posted percentage 10 times per second. Users can collect points at any moment at a rate of the current differential times the number of clicks they entered. For example, if you see at moment X that the market estimates Rand Paul’s chance at winning the Republican nomination at 20% and you click five times on “up,” then you could collect 50 points for closing your bet at moment Y when the estimate reaches 30% (50 points = 5 clicks x 10% differential).

PredictIt is a joint non-profit venture of Victoria University in Wellington, New Zealand, and the longtime political technology company Aristotle International. Here, users can buy low and sell high with real money, up to $850 per event. PredictIt creates “continuous double auctions” whereby a user can pair with another such that the sum of their probabilistic bets equals 100%.

Once a person estimating Paul’s chance at 20% finds another person who wants to place a marker down on 80%, a transaction can be recorded. At the designated end point (the nomination), winners collect their wager plus the profit pegged to the differential, less a 10% administration fee. So if you put down $100 on Paul at 20% and he wins, you collect $172, your investment plus 90% of $80. You can also buy, sell, and trade based on current offers in the market at any time.

The bet being placed by Pivit and PredictIt, of course, is that people want to capitalize on their sense of the future. Daniel Kahneman tells us that “we tend to exaggerate our ability to forecast the future, which fosters optimistic overconfidence ... the most significant of the cognitive biases.” Nate Silver warns us about bets cued by our partisan leanings and the media pundits we see and hear; the former flow from confirmation bias, our penchant for seeing what we want to see, while the latter emanate from faith in alpha types paid to be fast and assertive, not correct. 

Having read both men’s books, I don’t trust myself to play forecaster on either site. But if I could buy stock in either company, I just might, pending due diligence. (Disclosure: I know and respect John Aristotle Phillips, the CEO of PredictIt.) That’s because even though following politics closely is a minority taste, the instinct to gamble on the future of something well-known and much-talked about seems widespread, constant, and hard to resist.