I have set the personal goal of creating a forecasting bot by January 1st 2027 which can compete with at least 2 of the top 5 best forecasting bots on Metaculus. Resolves "Yes" if on January 1st, I personally believe that I have built a forecasting bot capable of that.

History: I am a Metaculus forecast by the name of Capital. Some notable achievements I have include:

On September 14th, I have started building a forecasting bot after the Metaculus Summer Cup had robots win both the #1 and #2 spots on the leaderboard.

This was the results of the last Metaculus Summer Cup.

On January 1st, I will estimate if my bot can compete with at least 2 of the other models above (Preseen, Cassi, Futuresearch, Mantic, Laertes) on a Metaculus Cup. If it can, I will resolve the question Yes. (If one of the bots above has a major issue eg. -300 score, I will not include it).

My bot does not have to outperform the listed bots above on the Fall Cup.

This is because my bot will get significantly better as time progresses, and the Metaculus Fall Cup is time weighted, meaning that the weaker performance of the earlier predictions will undervalue it. The resolution will rely on me estimating its performance based on a medium cost forecast ($2 or less) and whether that is competitive with top models.

I am willing to spend ~$200 total, and am not currently using any Metaculus/Asknews token subsidies.