I’d hoped for a more significant number (/ percentage) of MM users participating, but regardless, the number of NO voters is higher. Significantly higher than it was 24 hours ago, reinforcing @IsaacKing ’s Keynesian beauty pageant concern, but even then, NO was already slightly higher than YES prior to the recent run of no purchases in the last 12 hours.

In any event: I don’t think much about community consensus on this topic can be decided from this outcome; nor did my mind change very much. I think I’m less inclined, one week later, to be concerned about this sort of behavior in low grade background levels having a negative effect on MM as a whole. But not so much that I’ve abandoned any personal distaste for this sort of thing, either.

Thank you all for participating! 😄

I believe market creation cost should count as negative profit and recouped liquidity should count as profit.

@Yev That would mean market creators lose profits on all markets they create, right?

@Yev Wouldn't this disincentivise creating markets that get bet close to 100%/0% among people who care about their profit number? (Unless you also count unique trader bonuses as profit)

@a I guess unique trader bonuses should count as well.

I was thinking authors setting the initial probability should not count for or against their profit - possibly special case any author trades before anyone else trades on the market. This would address the specific case here and also address a fairly common concern in general - e.g. this was the main reason why the World Cup tournament created separate markets instead of using markets created by other authors - because the author gets an unfair profit advantage from correcting the initial probability.

I don't think counting liquidity costs and unique trader bonuses towards profit is a great solution - that would change the metric to be more like "net earnings including predicting and authoring" which is a different metric than "net earnings from predicting alone"

@jack You could frame it as the market creator being good at predicting what type of market will attract lots of traders.

@jack Liquidity costs seem to be counted as negative profits for free response questions. I'm at -63 for the one I posted despite not making any trades.

@StevenK Have a link?

@IsaacKing

@StevenK Interesting. That seems like a bug. I'll report it on the Manifold discord.

@IsaacKing Thanks!

See previous discussion on https://manifold.markets/Yev/what-of-manifold-users-think-its-so (which is a specific case of your question)

@jack I interpreted this market as being about markets that are resolved quickly, before anyone else has a chance to bet, while the market you link seems to be about markets that have already attracted traders.

@a Yes, that’s what I am aiming at. I can clarify the description if needed. A “sole participant” self-market seems to have little to offer anyone else beyond being fun as a joke and all.

@MattCWilson Got it, yeah, the description is clear but the title made me think it was a broader question. I would suggest adding the word "immediately" to the title.

I would say that on the question of immediately resolving a market you made, it could be kosher in some scenarios (e.g. you're testing something about Manifold), and in other scenarios it's not (e.g. you're abusing it to get leaderboard profits). I would imagine in most situations there isn't much legitimate reason to do it (I'm not counting immediately N/A resolution which makes sense if you made a mistake).

Also I want to note that the "immediate" part is important. I've made several markets where I asked a real question, I predicted, and the question duration goes by with nobody else trading just because they weren't interested, and then I resolved it as normal and earned a tiny bit of profits for my efforts. I think that is totally ok.

@jack Totally sold that yes, “immediately” is an excellent point to articulate. Personal “measured self” markets where you essentially predict your own habits are totally cool.

Title changes and edit log added. Feeling slightly bummed that I had had it in the description but that that evidently was not clear enough. 🙁

@MattCWilson It's fine, the market was reasonably clear and I don't think the apology you added to the market description is necessary at all! I wasn't saying that I bet on the basis, I was just clarifying the distinction since we were also talking about "sole participant" markets which isn't quite the same.

@jack All good, but thanks for the feedback. Trying to learn norms and do right in the community. 😁 We did get 5 or 6 other yeses and I'm putting that there for their benefit, in the event they want to change positions. (Of course, seeing as they hold the majority, and everything discussed here in the comments about voting profit not belief and all... maybe they won't 😆 )

Oh hey, another Keynesian beauty contest where people have no incentive to vote their true beliefs, and can turn a profit by not doing so.

@IsaacKing I think these markets are a fine way to generate discussion. Let people bet on the market for profit, and others discuss the question in the comments.

@IsaacKing sorry, new around here. Is there a way I could better have constructed this market, in your opinion?

@MattCWilson This market is fine as a way to create discussion as jacknaut said, but the probability it displays will be largely unrelated to the consensus. You've effectively just created a duplicate of /IsaacKing/will-there-be-more-traders-who-hold

If you want the market probability to reflect anything about the real world, its resolution needs to be linked to some real-world outcome that is correlated with what you want to know. You probably either want a poll or a pseudo-poll weighted by your interpretation of other people's opinions.

@IsaacKing Thanks for the thoughtful reply. I understood what you mean by “Keynesian beauty pageant”; I’m looking for something more constructive - like the remainder of your comment.

That said, what disincentivizes market participants from not carrying their preference falsification into the poll / psuedo-poll, just for profit maximization?

@MattCWilson I think I got a double negative in there… subtract the not. Er, wait, do I add a not to negate the existing not…?

@MattCWilson Social repercussions.

The norm on Manifold is that honesty is expected in comments, market descriptions, and discord messages, but it's always acceptable to place any bets you want to place.

@IsaacKing Ahaaaaa… now see, this is the sort of opaque-to-newbs context that I was hoping to get from creating this market.

Now I kinda want to create a KBP to test that claim, to see how far y’all are willing to ride that pony. Haha

@IsaacKing Also, wow, reading the comments on your linked polls, some of you folks have gone waaaaaaay deep on your pony riding. 😳

@MattCWilson Also makes me feel like, if this is truly “the culture” of MM, it could be really cool for the dev team to embody that in some more structured way in the app. Like being able to both place a bet but also have a position-free “belief” entry against the market, that defaults to the same thing you took a position on but can be flipped/changed if you wanted to signal that you are betting contrary to your own actual belief.

@MattCWilson Comments be hard to data mine, yunno? Also, would be hella more transparent to newbs, and maybe less likely to engender honor/integrity polls/markets?

To add to the above: even if people traded on their honest beliefs, holding YES doesn't mean you support YES, it means you believe the market price at the time you bought was lower than it should be. For example, if the market price was 10% and you believed 20% was the right value, you'd buy YES - but that doesn't mean you think the market should resolve to YES - you actually think it should resolve to 20%.

For this reason and the reasons Isaac discussed, I recommend using a poll instead. In this case, I would probably make the poll ask for a rating on a scale of 0% to 100%, and then resolve the average or median poll response. See my post https://manifold.markets/post/selfresolving for more on the topic.

@jack This is awesome. Thank you for the guidance - I had not found posts yet!

@MattCWilson Coming back to say this really has me thinking. Is there a way that MM could add some cheap, “good enough” detection of self-resolving markets, and offer up a suggestion of alternatives, to assist new market creators?

A pet theory of mine is that one critical bottleneck to prediction markets going mainstream is giving laypeople good tools for creating solid markets. Obviously there is no substitute for practice and experience, but useful tutoring/training guidance could expedite the process.

I’m thinking of something maybe along the lines of how Stack Overflow takes steps like checking for similarly-worded questions as just one way to help new users avoid one common pitfall. MM could have a similar check, but also have an additional check for phrases like “based on majority” or “most popular side” or something, as well, and pop a link to this excellent post.

(ps, unrelated: totally acknowledge that SO has a number of other questionable/problematic practices and isn’t an analogy I love here, but it functions well enough for this particular use case).

@MattCWilson Yeah, I think that would be great, and there have been suggestions along similar lines for other common market flaws - e.g. suggesting writing resolution criteria, suggesting adding some examples, suggesting picking a reasonable close date, etc.

@jack How exactly does one discover “posts” in the UX, besides a direct link like that? Not finding them under “Blog” in the main menu / app More button, nor linked from your trader profile page. 🤷🏼

@MattCWilson Hmm there used to be a bunch of places including the homepage and profile page IIRC, but I think posts were barely used so they are probably getting deprecated. I suppose you can get the same functionality by creating a market.

@jack I don't think that's very likely. Many groups have an About post.

@Yev By deprecated I don't mean removed entirely, I mean deprioritized and discouraged (e.g. showing them in fewer places than they used to)

I think right now the main place you can find posts is in groups - in a group page, click the posts tab.

@jack @Yev - found them; they occur at the top of the section, before the comments, on the comments tab of the user profile page. Not... the most intuitive location. I'd go with an optional additional tab when userPosts.length > 0

Might go push a PR 😁

@MattCWilson Woo! 😁 https://github.com/manifoldmarkets/manifold/pull/1364