Prediction Markets In A World Hostile To Gambling
By Ben Golden
Two key take-aways from the
emerging scandal surrounding Daily Fantasy sites: one, gambling data can be
extremely valuable, and two: the only thing Americans love more than gambling
is hating on gambling. Taken together, these findings illustrate why
large-scale prediction markets present a path towards improving human knowledge
in a wide range of topics.
DraftKings and FanDuel are the two
largest players in the Daily Fantasy Sports market.
If you're a sports fan and
haven't heard of these sites, then you've probably been living in a
their ads are all over sports broadcasts, the internet, podcasts, public
buses and trains, constant-running infomercials, any everything other than
the walls of caves
These sites allow participants to
wager money on a group of sports players' performances, and the
participant whose players accrue the most points win the pool.
While traditional sports-betting is
illegal, Daily Fantasy is allowed due to a
legal loophole created in 2006.
Allegations emerged last week that
employees of the two leading companies were using internal company data
relating to users' betting trends to play on each others' sites, and
had won large sums of money doing so.
All hell broke loose.
Americans' appetite for gambling takes a huge dip whenever there's a perception of unfair advantage--in this case, DraftKings and FanDuel employees being able to leverage inside information--the wisdom of crowds--while outsiders cannot. There are similar concerns with PredictIt, the political betting site. When these concerns gain some momentum, lawsuits are filed, investigations are formed, and laws are changed. In the words of TruthCoin founder Paul Sztorc, "these businesses (InTrade) close down due to interference, public misunderstanding (betting taboo)."
The flip side is that the intelligence generated by gamblers has real value. In this case, the crowd is generating insight into how players and teams will perform, which can be used (among other ways) to gamble on player performance. The aggregate of Daily Fantasy players' lineup decisions is a powerful forecast of player performance, and likely outperforms individual players' decisions. Of course, when only some individuals have exclusive access to the aggregate data, that makes it much easier for them to masses--hence the problem.
So the question becomes: how can we capture the forecasting information created by gamblers when we have such low tolerance for gambling? I see three possible solutions:
The public has a higher tolerance for
gambling when it's highly regulated, such as in financial markets.
However, in the United States, regulatory agencies tend to be highly
conservative about which types of gambling they allow, with little
consideration given to the information value that markets generate.
It's possible that a
cryptocurrency-backed prediction market solution, such as TruthCoin or Augur,
could enable betting markets to thrive on the black market. I'm
skeptical, but open to convincing.
- There's solid evidence that play-money markets, which are entirely legal, can generate forecasts that are roughly as accurate as real-money markets.
The third path seems most promising to me, since it sidesteps many of the hurdles (legal, regulator, financial, psychological) that prediction markets face. At Cultivate Labs, we just launched a new free prediction market for Sports, and will be rolling out additional markets for other topics shortly. We'll be closely monitoring our forecasts, and comparing them to other predictions. Who knows, maybe you'll even be able to use Cultivate Forecasts data to crush the competition in Daily Fantasy Leagues (assuming they survive).Ben Golden/@BenGoldn is an Engineer at Cultivate Labs.