16 Comments
User's avatar
Quroe's avatar

Even though my derivative market “Futarchy's Fundamental Fix” failed to fix the percentages, I think it was a fun attempt to science out a solution. If it didn't work, it didn't work! And now we know!

Expand full comment
dynomight's avatar

I also thought it was a good experiment. I actually tried to contact you before writing my post describing the experiments to see if you wanted to explain the theory, but couldn't find your contact info anywhere.

Expand full comment
Quroe's avatar

If you were asking about my contact info, sorry for the late response! I'm relatively new to Substack and I am still trying to figure out where all the bells and whistles are.

I make it intentionally difficult to find my contact info. Direct messaging me here, on Manifold, or finding my Discord profile through the Manifold Discord server are your best options at reaching me.

Expand full comment
Isaac King's avatar

If you have an account on Manifold you can message people directly on the platform from their profile

Expand full comment
Ben's avatar

No, I loved the idea! Sorry, I didn’t make that clear.

Expand full comment
Quroe's avatar

I never doubted that. I loved this post. 😊

Expand full comment
dynomight's avatar

Good post, you might also be interested in these markets (sorry if you already linked to them!)

https://manifold.markets/Tumbles/futarchys-fundamental-flaw-the-mark-d9LsnOnzhI

https://manifold.markets/Tumbles/futarchys-fundamental-flaw-the-mark-OR85qIqIcN

Expand full comment
Ben's avatar

Thanks! Ya, I saw those and was interested, but I couldn't quite parse the trading behavior and what to conclude based on it (partly because there were only about a dozen traders on the markets), but I wonder if @Tumbles has anything to say on the matter.

Expand full comment
dynomight's avatar

Yeah, unfortunately I think they both ended up being fairly inconclusive. For the first one, the coin was sadly revealed before the market closed. For the second, I think it's fair to consider it as weak evidence against my claims. Though let the record state that I preregistered that it should be considered weak evidence in advance, before the results were known! :) https://dynomight.substack.com/p/futarchy-market/comment/149593728

Expand full comment
Isaac King's avatar

> That we have perfect information about the odds of the two coins (ahem, candidates). This is almost never the case in real-world decision-making. We cannot know with absolute certainty that there is a 59% chance of our president invading Venezuela, say.

BRB, reporting you to the Manifold admins for frequentism

Expand full comment
Ben's avatar

Ahahaha, ya I guess I could have phrased this more eloquently. Like, if I know that there's precisely a 59% chance of something happening, I can precisely quantify the risk in betting on it. Like, there's no adverse selection or secret information, or whatever, so I can feel confident in my betting behavior in a way that's impossible in most markets.

Expand full comment
Isaac King's avatar

Yeah I think the real distinction between an election and a coin flip is "can other people learn significantly more about this event than me?" Some related discussion:

https://manifold.markets/IsaacKing/best-explanation-of-how-some-probab

Expand full comment
Isaac King's avatar

> Robin Hanson swears up and down that market manipulation is not an issue, but I think this is just completely false. A recent study, carried out on Manifold(!) supports my conclusion, as does common sense.

The study only included tiny markets that cost less than 200 mana to move, so I think it's a stretch to call it support for this claim. They found that larger markets within this range were harder to move, exactly as Robin would predict.

Edit: Oh, they also tested sweepcash markets that cost up to $175 USD to move, and found only slightly less susceptibility to manipulation. That does seem bad.

Expand full comment
Ben's avatar

Yes, this is true, but Hanson has made the case that not only does market manipulation not work, but it also makes markets *stronger* because it provides an incentive for finance.

I think market manipulation *does work*, but becomes harder and more expensive given larger markets, which I think lines up with what this study finds... and like, intuition.

Expand full comment
Isaac King's avatar

> it also makes markets *stronger* because it provides an incentive for finance.

This could be true in the long-term, but it's surely false on the particular market being manipulated. (Fails the drunk mormon heuristic.)

Expand full comment