Project summary

Many high quality questions are created on a number of different forecasting platforms. Mirrorbot allows the Manifold community to benefit from this by creating and managing mirror markets for these questions.

Besides being a source of high quality markets, mirroring makes it easier to compare performance between platforms. It also allows traders to arbitrage across platforms, and to be confident markets will resolve correctly and on time.

I started working on Mirrorbot late September this year. Since launching it in October it has created 150 markets attracting 1600 traders and 30 followers.

Mirrorbot: https://manifold.markets/mirrorbot?tab=questions

Github: https://github.com/jkerkhoff/mirror_bot

What are this project's goals and how will you achieve them?

Since the bot already exists, the goal of this project is to further improve it. Ongoing development is likely to focus on

1. Additional platforms. Mirrorbot currently mirrors questions from Metaculus and Kalshi. Next on the list would be Polymarket. Insight would also be nice to have.

2. Multiple choice markets. Right now only simple binary markets are supported, which is quite limiting. Besides making it possible to mirror multiple choice markets from source platforms, for many ongoing/grouped markets ("Will the fed raise rates before x date") unlinked multiple choice markets would be a good fit.

3. Other features. For example, allowing people to ask the bot to check if a particular market should be resolved (right now it only checks every 4 hours), extending support for managram mirror requests (currently only Metaculus is supported), and trader bonus sharing for requested mirrors.

How will this funding be used?

Funds will be used to reward the time I (and potentially others) spend on Mirrorbot development. Work so far has been mostly motivated by personal interest, but there are other projects I could be working on and some of the project's goals are not super exciting to work on (the Polymarket API looks kind of painful to use, supporting multiple choice is going to involve a bunch of refactoring and glue to support different schemas).

Who is on your team and what's your track record on similar projects?

Right now the team is just me (though @kenakofer has contributed in the past, adding Kalshi support). If anyone else would be interested in contributing, feel free to contact me on here, Manifold (jskf), or on Discord (jskf_).

As for my track record on similar projects, I am the creator of Mirrorbot ;).

What are the most likely causes and outcomes if this project fails? (premortem)

• Supporting multiple choice markets well will require some additions to the Manifold API, though it seems some of these are already being worked on (editing existing markets being the main one).

• I'm currently unemployed. If I get a new job I will have less time to work on this project (and might have to get permission to do so at all).
  

What other funding are you or your project getting?

I got a 15k mana bounty from @Genzy. Austin also contributed 20k mana to Mirrorbot's bankroll to fund market creation. I haven't received any cash funding and have previously been described as the worst paid developer on Manifold (by me).