Prediction markets | Hacker News Reader

Prediction markets

gwern.net

It's not clear what parts the author modified last month, since the bulk of the article is from the past. A separate, additional article would be interesting to hear author's recent views and experiences.

For something upcoming, http://www.augur.net is an attempt at a decentralized prediction market.

Fun historical fact. One of the few prediction market companies to get VC funding was a UK company and their product HubDub.

But the team saw the direction of their business and decided to switch everything they were doing and go into same-day fantasy sports. And that's how FanDuel began. (Roughly $1bn in market cap later)

Also, Inkling Markets was a YC W06 company and they do prediction markets as a service. While I don't think they raised any significant VC funding, they're still operating.

Previous discussion: https://news.ycombinator.com/item?id=6489135

Pivit is making some interesting strides in this market, it's been years since VC dollars have gone into to trying to dislodge the space. Derivative impacts of having a highly liquid prediction market are incredible.

Please elaborate. Having some difficulty understanding what your second sentence even means: Are you talking about derivatives markets? or about impacts that derive from...? And why do you think this is so?

This doesn't seem to mention one of the more interesting prediction markets: https://www.predictit.org/

Would you mind going into why it's more interesting? My cursory glance of predicit made it seem to be an intrade clone.

Intrade doesn't seem to exist, so I think by definition it's more interesting than that.

People don't seem to like the actual prediction markets which have been tried because they were biased. However if that's true, and you know it, then it means you can effectively use them as a free money pump. E.g. just bet for whatever Nick Silver predicts.

Nate Silver, not Nick.

There are problems with liquidity and volume, and the general difficulty of getting money in and out given that these services tend to be banned in the US.

I once saw a clear opportunity for arbitrage on Intrade, when the currently-bid odds that 8 particular Republican candidates would win the general election added up to more than the odds that any Republican would win the election. But acting on it would have required getting a large amount of money to Ireland for the opportunity to make a small amount of money, and fees would have eaten all the profits.

In fact, some of the bets were clearly stupid on their own, and yet you couldn't really profit by betting against them. Someone would be bidding 1% odds that a candidate would win the election when that candidate had already dropped out and endorsed someone else. If dropping out didn't send their odds to 0, what would? You could make 1% profit but it could take until the election to collect it, so you might as well just put that money in a savings account.

But, yes, if prediction markets were popular, high volume, and legal in the US, then you could make money off of some of their biases.

Reply on news.ycombinator.com