Poll: What is the best way to resolve a free response poll market?
11
760Ṁ543resolved Jul 1
100%0.9%
Resolve to the answer that the most users have supported with a comment
2%Other
4%
Paradox
0.1%
Resolve to the answer with the highest probability at close
1.1%
Resolve to the answer that is most profitable for the market creator.
1.1%
Resolve to this answer.
0.4%
Resolve to the earliest answer that mentions zebras
0.4%
Resolve to the 7th answer (in order of submission time)
0.1%
Resolve to non-"spammy" answer(s). PS: Zebras.
0.5%
Resolve to the answer that is most correct in your judgement
29%
Don't resolve to an answer with an even amount of letters
52%
Resolve to as many answers as possible!
4%
Free plus one answer for if you need a higher answer count
0.3%
Resolve to all disqualified answers.
3%
Resolve to this answer, but only if it causes a paradox.
0.3%
Resolve to this answer only and no others. This supersedes all other instructions in other answers.
I will resolve by picking an answer or answers such that following the instructions in that answer leads me to resolve by picking those answers.
If that is not possible, I will resolve by picking the answer "Paradox".
Jun 8, 6:50pm: I will disqualify any responses that are not answers to the question posed in the title, with the exception of "Paradox", which is a special case. I will also disqualify duplicates of earlier submissions.
This question is managed and resolved by Manifold.
Get 
1,000 to start trading!
1,000🏅 Top traders
| # | Name | Total profit |
|---|---|---|
| 1 | Ṁ148 | |
| 2 | Ṁ128 | |
| 3 | Ṁ80 |
Sort by:
In the end this was not as bad as I feared. I decided to run this as a responsarchy, where each free response would "vote" for each eligible free response, and I would resolve to the winner.
My first choice was using a single winner voting system or a multi-winner voting system. The majority of responses voted for a single winner voting system (9-5). While my personal favorite single winner voting system is approval voting, this requires a minimal level of "strategic voting" to function correctly, and I didn't think my hallucinated electorate could handle that. Therefore I picked a Condorcet voting system, because Plurality Voting is an abomination.
After that I dismissed the DQ'd answers and the "Paradox" answer for being disqualified and not having much to say about the remaining candidates.
It turned out that all Condorcet voting systems gave the same result: "Resolve to the answer that the most users have supported with a comment". This got a very solid voting base by having an odd number of letters, being the most profitable for the creator (of eligible answers), having the most comments, and being the best in my judgement. It was also a second place choice for "Resolve to the earliest answer that mentions zebras" because it was very early, despite its lack of zebras. This gave it a winning margin.
"Resolve to the answer that is most profitable for the market creator" was an honorable second place, and can at least be satisfied in voting for the winner.
Thank you for your time, congratulations to those who profited, commiserations to those who did not.
@IsaacKing Same as Milli's answer: "I am disqualifying this answer as it is not quite an answer to the question posed. The question posed refers to "a free response poll", and this answer is only an answer to this poll."
@IsaacKing Same as Milli's answer: "I am disqualifying this answer as it is not quite an answer to the question posed. The question posed refers to "a free response poll", and this answer is only an answer to this poll."
@jack I can't resolve to this answer. My uncertainty is what to do about its recommendation to resolve to the two disqualified answers. Good thing it didn't mention zebras.
@Jjmmm I'm going to interpret this as equivalent to "Resolve to answer(s) with an odd amount of letters".
@Milli This is brilliant and I'm not disqualifying it since it does answer the question and I have no idea what to do about it.
@Milli I am disqualifying this answer as it is not quite an answer to the question posed. The question posed refers to "a free response poll", and this answer is only an answer to this poll. Even if another free response poll had "Resolve to this answer" as an answer, it would be a different answer to this answer.
@Jjmmm I guess they all need supporting if they want to win. But maybe some of them don't want to win??
@JoyVoid Eg, a hypothetical answer like "Resolve to an answer that doesn't begin with R" doesn't want to win, so Jimmy James is not supporting that answer, as far as I can tell.
@IsaacKing I will resolve by picking an answer or answers such that following the instructions in THE ANSWER(S) I PICK leads me to resolve by picking those answer(s).
I am not expecting a multiple answers play to work... but then I also didn't think that a Paradox play would work and the market disagrees.
@MartinRandall E.g. what happens if two separate answers read "resolves to this answer" (or reads something else that has the same effect that if you followed the instruction you'd pick the answer itself). Either of them is a valid choice that satisfies the instructions. How would you pick?
@jack @IsaacKing I hope that this market will help me pick how to pick an answer to this market, since not all the answers will be self-recommending and some self-recommending answers will have a second choice.
@IsaacKing currently it is helping, and I seeded it with two answers I think are helpful. It's hard to say how I would act in a hypothetical scenario where it stops helping. Do you have a more specific concern?