51 first dates, 16 second dates, and 9 third dates
Manifold Love's first month of tracking dates + the launch of OkCupid-style compatibility questions + what's next
I'm excited to share with you Manifold.Love’s launch stats, recent site updates, and our future direction.
This month, we successfully doubled our dating profiles to over 1,000!
From this set of early adopters, worldwide, it is really cool to say we have created real impact in producing the following dates, recorded through the resolution of prediction markets:
51 first dates
16 second dates
9 third dates
Congratulations to those who participated — I hope these dates were enjoyable and are leading to greater fulfillment.
Note that these are just the user-reported results (since Nov 9th), which could be slightly inaccurate. I know a couple cases of fake resolutions (where we banned the user who reported incorrectly) and of users who did go on a date, but for whatever reason did not have the confidence to announce it publicly by resolving the market YES.
A user shared on our Discord that they're going to Greece with someone they met on Manifold Love. Sounds like a blossoming manifold love.
Checkout also Scott’s writeup on Manifold Love after one month.
Compatibility questions!
Next, I’m excited to announce Manifold Love has implemented compatibility questions (and now computes compatibility scores with potential matches).
You asked for it, and we built it!
Many independent users have approached us to say that our site could be improved with a set of features similar to the original OkCupid.
After building it, I’m glad that we did (feel free to resolve the relevant market).
When three new users in row signed up and immediately answered 100+ questions, I knew that this was going to be a hit. (h/t to Eli Tyre for manually adding ~150 compatibility questions!)
What’s next for Manifold.Love?
Since launching our Minimum Viable Product on October 26th (coded in just two weeks! 😳), our product has gotten a lot better.
We reworked the prediction markets to track dates, send emails when something happens, polished the onboarding flow, made your search filters save, added a browse mode, added compatibility questions, and more.
But I don’t feel that this is even a version 1!
In addition to missing a mobile app, there are two problems to solve in my mind for would-be daters.
It’s not that easy to express interest.
Some users are messaging each other instead of creating a match which would set up a public prediction market. They do this because they don’t want to make such a loud proclamation of their interest.
We also want to have one clear moment to start messaging each other and ask for an in-person date. Today, that’s mostly lacking.
Possible solution: add likes!
I think the standard action of tapping a heart next to someone’s profile to show that you like them will help a lot. If you both like each other, that’s a clear moment to ask someone out.
And, what if matchmakers can also like potential pairs… then we can sort matches by total likes from all friends and strangers. That will be amazing.
The prediction markets are complex and not that actionable.
We started with markets on your chance of a six-month relationship, and switched it to four smaller markets that decompose the six-month relationship into four steps, each one conditional on the previous: first date, second date, third date, and six-month relationship.
While it’s kind of cute to see people’s relationships progress, I’m not sure how actionable the markets are.
For one, there’s not much trading. But at least part of the problem is that after the first date, the daters themselves know so much more about their compatibility that any bets from outsiders are pretty much irrelevant.
Possible solution: One market on chance of third date, conditional on messaging each other.
To be useful, then, we need to focus the markets earlier in the search process. They should help you choose who to message in the first place.
We can even close the market after users both message each other, because there’s no point to trading after the die is cast. With this change, traders will never lose because of insider trading.
Additionally, we want a structure where users can pay to subsidize their markets to provide more liquidity. This incentivizes traders to put more effort into correct matchmaking because they can earn a larger reward.
If we change the question from “Will we go on a third date?” to “Who will I next go on a third date with?” then there is exactly one correct answer, and this allows us to subsidize all proposed matches without losing mana to poor matches.
(Bonus) Matchmakers have trouble keeping track of their friends within the app
Involving matchmakers that are not themselves trying to date is a very unique aspect of how Manifold.Love works. It can be a lot of friction to keep going back to search for each next friend to click into their profile and see how they are doing.
Possible solution: Create an “Add friend” button and surface friends first on the homepage and compatible lists
If we let users save their friends, and show them first on the list of profiles and with the browse compatible users lists, then it will be much easier to check in on them and propose new matches for them ❤️❤️❤️
Manifold Dating Show?
We’ve another secret project in the works — it’s called “Bet on Love”!
If you’re interested in participating as a contestant, add yourself to this market!
That’s all the news for now!
Manifold Love is already a strong contender for the best open source dating app. (Yes, I’m calling it now: Manifold Love is awesome.)
But I can’t wait to get in a few more iteration cycles, throw some ad dollars behind it, and scale our app to take on the giants. Ultimately, we can show the power of crowdsourced matchmaking via prediction markets to produce amazing relationships.
—James
It seems like maybe what we want is to allow users to submit any question they want about a pair of daters, and then we can see which kinds of markets attract traders. Instead of trying to figure out what the right market structure is top-down, just let a thousand flowers bloom, and let people create the markets that are most interesting to them.
This seems like it might be a moderation nightmare. But, maybe the daters themselves have to approve the market, before it goes live, for decency reasons.
(Maybe in the longer run, daters can have a dash-board of what kinds of markets they're comfortable with. ie "Yes to questions about who I'll be in relationships with, but no to questions about one-night stands". This could have some tasteful defaults, and people have to opt into there being markets about what sex-acts they'll engage in. )