Starting off with 5000 in the pool, we can add more if it seems necessary. I can try to beg SirSalty for 50000 more if this gets popular.
The main goal: I want to learn about what manifold users consider to be well-run markets.
Sub-goal: I want to compile as much information as possible about this to share with the powers that be.
How to earn some mana: Share markets where the creator did an exceptional job of running the whole show. Maybe they wrote really good criteria. Maybe they did a good job of selecting what options to use. Maybe they were masters of the comment section.
I'll give you a little bit of mana if you share something basic (~10 mana).
If you explain in detail what the creator did well, I'll give more (~100 mana).
If you can explain deeper concepts you learned from the market and show how you incorporated those ideas into your own markets or how they spread, I'll give even more again (1000+ mana).
If you have a general concept that you learned and really liked, that is not attributable to a single market but is a broader pattern, share that also. Just link to at least one market as an example.
Don't report the same thing/person someone else already reported unless you have something new to share. Just comment below existing entries and/or upvote them.
1,000People are also trading
/Lunar/will-more-than-1327999-acres-burn-d
That's a currently running one, but there were 2024 versions as well.
Why exceptionally well-run:
There are resolution criteria
Resolution source is provided
Additional information is provided on how and when the source usually updates
Historical data is provided
Market is updated with new numbers
There is a clear disclaimer be the creator on whether he will trade himself
There is a clear disclaimer by the creator on how he will handle clarifications (temporarily close, later reopen, if necessary)
There is a clear disclaimer by the creator on comments (do not count as clarifications, unless they are added to the description)
Here's part of the disclaimer:
If Any Clarification Is Needed, I May Temporarily Close The Market To Make Clarifying Statements & Than Re-Open ; [...] Comments are not a clarification unless posted into the description.
The only nitpick I can find: A commitment to @traders when the description changes in a relevant way (e.g. by a clarification). Not sure if the creator does this anyway.
While this is exceptionally well-run, a clear resolution source does most of the heavy lifting. Just the title plus "resolution source: xyz" would in this case already suffice to make the market way better than average.
https://manifold.markets/Joshua/will-eliezer-yudkowsky-win-his-1500
This is one of my favorite markets, but not because of the actual prediction. This is my favorite measure of discount rate.
This market:
has high volume
true probability is, imo, over 99%, and I think that's what something like Metaculus would say, but there's a non-zero amount of dissent to that. markets with uncontroversial foregone conclusions tend to be less affected by discount rate because someone who isn't savvy will bet it up to 99%/1% and nobody will bet it the other way. with genuine dissent, there's opportunities for market making, so there's action at 92%-93% even from people who think the true probability is higher
will have a resolution that is likely to be uncontroversial (the underlying "will UFOs have a worldview-shattering origin" may be controversial, but "will Yudkowsky win the bet" likely won't be)
https://manifold.markets/tftftftftftftftftftftftf/can-i-go-2-weeks-after-the-election the blueprint silly market.
it was a question that was hot/popular topic adjacent (arguably the election was THE HOTTEST) which I felt is a cool approach. it invited lots of fun speculation. the creator uploaded entertaining youtube vids during the market, logging her progress (and showing dedication). The comments kind of helped me first learn that on manifold there are a large group of 'characters' (or personalities) in a way, and they were visible here. It enriched my sense of community here.
This is more in the "general observation" category, but I appreciate it when people acknowledge the subjectivity of a market using "will I think"/"will Manifold think" in the question wording, rather than using objective language like "will an LLM be able to...".
Another category of market is "default no" vs "reasonable" - these are the ones where the resolution criteria appear clear if somewhat subjective, but then it turns out that it won't resolve unless the market creator is provided with exhaustive evidence that there exist no possible counter-examples (as opposed to merely showing something is likely or unlikely).
A notable example where this tripped up a number of people is the one about whether an LLM can draw a pentagon. They tried to be explicit with criteria for the prompt, but the standard of evidence for judging that it happened was so extensive that despite being provided with many examples of a LLM-generated pentagon, the market wouldn't resolve because sometimes the model wouldn't do it.
A market that at the outset required effort from traders, but with the creator themselves making significant initial and further ongoing effort and engagement during the run of the market to follow through solving the random number generating problem that surfaced.
https://manifold.markets/inaccessibles/will-a-randomly-generated-integer-f
@ian submitted this one but I can't give him a bounty because it's not in the comment section:
/JRR/if-trump-wins-the-2024-election-wil-877b62a380ec
It looked too long for me to read it so if someone wants to earn a bounty you can explain what part was exceptional.