There is currently no real option for regular people who want to start prediction markets on dating success probabilities. There are those who suggest existing platforms can be used for this purpose, but those platforms don't work, as demonstrated here:
To make a small subset of the difficulties explicit:
Any market I create would have extremely low trading volume, and subsidies wouldn't help this problem much (and they'd fuck up price convergence by giving tons of money to the AMM anyways). So the results would mostly be noise/the opinion of the first few betters.
Having to pay 25c for suggesting partners is enough of a barrier that in the median case my market will have like ~three answers, like Sinclair's (at least at time of writing). Adding people I'm interested in to my own market is time consuming and also kind of awkward so I'm not gonna do it.
(Noticing confusion about this) AFAICT the vast majority of liquidity is wasted. On Manifold's current system, all of the choices get their own AMMs with independent liquidity pools. So let's say I put 1000$ up and people suggest 100 possible dates (and in the best cases I'd like people to suggest even more), the subsidy will get split 100 ways. This means if I only go on a date with the top suggestion, I'll really be getting 10$ worth of info on the option I actually went with, which is unacceptable. If this is incorrect feel free to contradict me in the comment section.
Since nobody else is publishing similar markets, that makes it a lot harder for market makers to justify spending time or energy to develop good general alpha or to spend time analyzing what makes marriages work, since I'd be the only person asking these questions. You need lots of people doing the same thing in order for it to be worth it to develop empirical models with any real backing.
Market makers lack lots of critical information about me, at least in an appropriate, regularized format that had the advantage of a dataset of previous relationships, which they could use to analyze success chances. Prediction markets aren't magic, manifold users aren't going to dm me relevant Big5 questions and run regressions with them over OKCupid data in order to get a bunch of internet points, they're just going to use what I write in the description and make a 30-second judgement call, which is not really what I want.
And this is only a sample of such problems. You'd be very foolish to suggest this platform currently supports this use case, much less has debunked it!
If, on the other hand, manifold.love or another site manages to solve a portion of these these problems to such a degree that I seen the site "usable", i.e. I can get actionworthy information out of it without lots of effort and for a reasonable price (<1000$) as a normie, then I will resolve this market YES. If not, I will resolve this market NO.
1,000🏅 Top traders
| # | Name | Total profit |
|---|---|---|
| 1 | Ṁ342 | |
| 2 | Ṁ80 | |
| 3 | Ṁ63 | |
| 4 | Ṁ39 | |
| 5 | Ṁ6 |
People are also trading
@MilfordHammerschmidt manifold.love has been out for a little over a month now. Does it address what you were going for in this market, to any extent?
@chrisjbillington Sold my position to make this statement. My current impression is no, based on the problems I outlined in the description:
"Any market I create would have extremely low trading volume": Surveying people with my demographics, this is certainly true on manifold.love. Male markets for non-celebrities get virtually zero traders.
"Having to pay 25c for suggesting partners is enough of a barrier that in the median case my market will have like ~three answers": Actually this was too optimistic. Most dudes get zero suggested matches. The women mostly get matches that the dudes suggest for themselves. Those match markets that the dudes suggest get almost no outside trading. So the markets don't exist, and when people create them they give virtually no usable information.
"AFAICT the vast majority of liquidity is wasted" - I ~think this is solved by Manifold printing enormous amounts of mana for every market.
"Since nobody else is publishing similar markets, that makes it a lot harder for market makers to justify spending time or energy to develop good general alpha" - Getting better over time.
"Market makers lack lots of critical information about me, at least in an appropriate, regularized format that had the advantage of a dataset of previous relationships, which they could use to analyze success chances" - Still true, unfortunately, only a couple dozen markets have "resolved" so far, and I don't think people have written any bots.
@RobertCousineau Furiously iterating on a prototype! Hope to launch something tomorrow (though truthfully it needs another few days).
@JamesGrugett my partner's said she'll join (along with me) when it comes out. I can't promise she'll be a MAU though unless it's good :P
[Redacted, duplicate comment]