Will Manifold fix the problems around incorrect market resolutions by the end of March?
11
44
230
resolved Apr 3
Resolved
YES

The two main problems right now are:

  • Honest market creators can resolve a market incorrectly by accident, and have no easy way to fix it.

  • Dishonest market creators can resolve a market incorrectly on purpose in order to turn a profit, and face next to no repurcussions for it.

These are both very serious problems that impact users' willingness to trust markets and put money into the site, and they're pretty simple to fix, so it's unclear why Manifold has not done so yet. For example:

  • Allow creators to change the resolution of a market within a few days of its resolution.

  • Give the community a way to flag dishonest creators, and display that warning prominently to others. (We used to have this, and Manifold removed it for some reason.)

  • Allow either Manifold admins or some sort of community consensus to re-resolve or N/A a fraudulent market.

There may be other approaches as well, these are just the most obvious ones.

This market resolves positivly if I feel the issue is at least mostly resolved for future markets. (Past markets don't need to be fixed.)

Get Ṁ200 play money

🏅 Top traders

#NameTotal profit
1Ṁ160
2Ṁ146
3Ṁ55
4Ṁ34
5Ṁ16
Sort by:
predicted YES

Looks like Manifold has now changed their policy and are allowing fraudulent markets once again. Neat. Wouldn't want us to become too trusting I guess...

https://manifold.markets/Hyperpolitan/will-a-meteorite-hit-my-house-by-mi

predicted YES

I think this should resolve YES. Any disagreement?

predicted YES

Has anyone recently encountered a market that resolved incorrectly and was not fixed, or was only fixed after significant effort and annoyance?

@IsaacKing I sure wish there was a process native to the site, but nevertheless it seems like admins are willing to do all the necessary reresolutions!

Agree, I consider the problems largely fixed!

bought Ṁ20 of YES

Allow either Manifold admins or some sort of community consensus to re-resolve or N/A a fraudulent market.

This looks like it happened here: https://manifold.markets/serpa/will-the-philadelphia-eagles-win-th#XbIiGmuiYgNxEanesdW1

What part of the condition do you think isn't fulfilled? Is it about the difference between mistakenly resolved and fraudulently resolved markets?

predicted NO

@StevenK We still don't have any way to flag dishonorable creators, nor fix previous markets.

predicted YES

@IsaacKing Does a solution require flagging dishonorable creators even if dishonorably resolved markets can be re-resolved? "Past markets don't need to be fixed" according to the description.

bought Ṁ22 of YES

@StevenK Traders need to be safe betting in markets. If they can trust that any incorrect resolution will likely be fixed, that's good enough.

bought Ṁ10 of YES

@IsaacKing I think then it comes down to how willing the admins will be to use this new feature consistently.

I'm pretty sure they're actually not simple to fix - I expect there's a substantial amount of technical work needed to be able to change a resolution after a market has already resolved.

Give the community a way to flag dishonest creators, and display that warning prominently to others. (We used to have this, and Manifold removed it for some reason.)

It was removed because it didn't work, with the intent to rework it, but at this point that doesn't seem to be happening soon. The problem with it was the way the thresholds were set up made it impossible to avoid having both false positives and false negatives. There are easier solutions like having the admins flag dishonest creators, but I guess that forces the admins to make more potentially controversial decisons.

Relevant markets but for 2023:

bought Ṁ40 of NO

@IsaacKing Would just restoring the "Give the community a way to flag dishonest creators, and display that warning prominently to others" feature be enough to resolve this market as YES? That only seems like it solves the "dishonest market creators" problem, not the "honest market creators" problem.

@Gabrielle No, we'd also need a solution for the honest mistake issue.

bought Ṁ10 of NO

@Gabrielle yeah I think that is addressed in the final sentence as it would probably need atleast one fixed and another somewhat fixed to resolve yes