Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
PredictIt is shutting down after CFTC withdraws no-action letter (predictit.org)
28 points by paulgb on Aug 4, 2022 | hide | past | favorite | 17 comments


This is such a bummer, but not shocked they're out-of-bounds given the bullets in the letter and how big they've become: https://www.cftc.gov/csl/22-08/download


Good luck to the crypto prediction markets, we need them more than ever now. https://polymarket.com/


No, we don't. They're illegal, and legal alternatives exist


[flagged]


No, I'm not. I'm the "I don't like multiple countries' worth of wholly unnecessary pollution to support an unregulated securities casino and its billion dollar fraud of the week" guy.

In general I have a hard time writing the word "terrorist" without sPoNgEbOb CaPs. They have less of an impact on the world than almost any other threat to my life. I'm genuinely more scared of regular things like car accidents and losing my wallet.


For people looking for alternatives, Kalshi (YC W19) is a regulated exchange on event contracts. It offers some markets similar to PredictIt (and more).


Why even look for alternatives, prediction markets suck due to inherent limitations in liquidity, making it impossible for them to ever fulfill their purpose of broadcasting what "the market" believes. One market participant with slightly larger size moves the state on low volume.


I’ve made hundreds of thousands of dollars on predictit. It was a great way to make a straightforward side income for me for many years.


hundreds of thousands across many years

this suggests to me that you needed to spend that money periodically as it was made, for your lifestyle, and/or that there was a maximum trade size for possible (or just for your risk tolerance)


They could be high volume though... There might be traders who run AI bots to scour Twitter and other news sources and do high frequency trading to earn money on volatility. Just like stocks and shares.


> There might be traders who run AI bots to scour Twitter and other news sources and do high frequency trading to earn money on volatility. Just like stocks and shares.

Yes, thats the point and promise of all prediction markets at any pitiful size.

I'm not seeing how this response addresses the limitation of them. They're not big enough and have had decades to subvert that. Instead its just stupid binary questions for events that frequently are not binary, the questions/markets are unilaterally posted with no analysis of whether it is a binary-enough event. Liquidity is fragmented, there are no market makers helping - from what I can tell - and market makers shouldn't appear with the broken structure of the markets and questions and arbitration of outcomes to begin with. And then, the profit is limited to just the binary option maximums, compared to an actual option with unlimited upside. A trader running an AI bot should just stick with stocks and shares and commodities then.

Prediction markets do not succeed well in monetizing events. They're a broken half measure.


"The Iowa Electronic Markets is a futures market run for research and teaching purposes. Traders can buy or sell real money contracts based on their belief about the outcome of an election or other event."

https://iemweb.biz.uiowa.edu/


While groundbreaking for its time (started 1988!), the IEM is a very poor substitute for PredictIt. While I could offer a half-dozen reasons why, some of which were noted in CFTC Letter 14-130 for PredictIT in the first place, it'd probably do to point out that their "current" markets consist entirely of:

* 2022 French Presidential Election Markets (I haven't looked at these too closely that maybe there's a reason they're still listed as "current" on the IEM site, but wasn't this election back in April?)

* 2022 U.S. Congressional Control Markets (3 Markets: Senate, House, and Overall)

If you're trying to use prediction market prices as a gauge of the state of the world, IEM is missing dozens of countries, does not give visibility into US state-level elections, and does not include the result of primaries or political events besides elections that nonetheless impact elections. The 2024 US Presidential election is not even on the radar yet there.

Credit where credit is due for the trailblazing, they're worth a look when they do have a market on an event going, and it's notable for being sort-of a proof of concept for prediction markets. But while IEM may now be the best fit within a particular niche in light of what's happened to PredictIt, it's not in the same league.


I liked them until they started emailing me for my SSN... I'm looking for an alternative now


If your winnings exceed a threshold ($600 IIRC?), they need this to issue a 1099MISC. Any IRS-compliant service would do the same, including PredictIt.


This is a real shame. It was a lot of fun to see how the markets behaved every election year


Betting odds perform much the same task, but with a substantially higher fee...


With markets weighing heavily against the current administration, of which cftc is a part, this is a no-brainer.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: