🏅 Top traders

Wait, didn't this happen after market creation? They have to be tagged just grab bag now:

@nikki were you aware of this when you bet no a day ago?

Gonna resolve this yes if no objections, and that's the last of them.

I think this became more of a wishlist of changes than things that were actually going to be updated by that date. I would recommend sharing this with site leadership to be actioned on.

@NFL Yeah lots of good suggestions here!

Without using any of the sort options, I would like to create a default order as to how the questions appear to people visiting my market. The user can then sort if they so desire.

There's no way this many changes would be shipped in the next two weeks. I think there's a lot of profit to be made just by betting NO on all the options.

On second thought, the one way I could be wrong is if improved prop bets markets are the big pivot that the Manifold team has in store...

good thing answer add notifications were restored or I might miss the real gems like this

@shankypanky Oh geeze

@shankypanky That's why I've spent years of my life fighting for a comeback of this feature 😉

@Lion thank you for carrying on the fight in my absence to ensure I'm always in the loop 🫡 immediate payoff obviously

Hasn't this already happened? Back on the 12th there was just the input box for searching/adding, then they hid it behind a "search" button, and now I see you can unhide it either with the "search" or "add answer" buttons.

@BrunoParga Yes true!

This would be superb to feauture certain questions and too keep the spirit of the market

@Lion Yeah there have been many times I have wanted to do this, like a debate market with a "who will win?" Pinned to the top and then assorted trending props below.

@Joshua Not necessarily a problem, but I do think that’ll extra incentivize people to be the creators for prop bet markets, and pin their markets at the top for the best visibility. That’s probably fine, you do want creators to control their markets, but if you already have people competing to be the creator in charge of a popular prop bet market, this will make it even more desirable

@Ziddletwix Should I add "Nationalize Prop Bets" as an option? 😅

@Joshua I mean tbh I don’t think some form of that would be crazy, especially when manifold.markets/tv is involved, which basically makes it “official”.

But it’s hard to get the details right. Running props is hard so you still need to offer SOME incentive, it feels bad if Manifold steps in last minute to take over something that you built, etc etc

@Ziddletwix Yeah it's a very complicated set of incentives. There is definitely a good argument that bet creators are under-incentivized right now because they're doing a lot of work running the market while users can just add options and sit back to collect trader bonuses.

Given how many subjective cases there inevitably are with these markets, I think one direction you could go in is to pay the market creator more in trader bonuses and let them officially appoint co-resolvers to help resolve, sharing the increased trader bonuses with those co-resolvers.

But then after you do that, you encourage a norm that the resolvers commit to not trade in the market themselves so that everyone can trust them to be unbiased.

I haven't thought about this very hard though, definitely a lot of ways you could improve the incentive system.

@Ziddletwix I feel like the people who compete to be in charge of managing prop bets market always is and will be significant smaller than for normal easy to manage & resolve questions like 'Who will be president?' or 'Who will be POTY?'. I don't think that this is a specific prop bets market problem.

This would be so nice, it's really hard to find 1 option out of 100 to resolve it

@Joshua On desktop it’s manageable, but on mobile forget it.

This is already how the "Trending" sort works.

See https://github.com/manifoldmarkets/manifold/blob/d9adce66d59fda87395a158f5efa1d949a29d87e/common/src/contract.ts#L73 and https://github.com/manifoldmarkets/manifold/blob/main/common/src/answer.ts#L121

  { label: 'Trending', value: 'liquidity' },

and

      } else if (sort === 'liquidity') {
        return 'subsidyPool' in answer ? -answer.subsidyPool : 0

@Gabrielle I'm confused—that's not how "trending" works when searching for questions right? (IIUC that's a measure of increase in popularity within the last ~week, right?)

Edit: actually I'm extra confused. when you sort by "Trending" in indepdendent MC, you clearly see a question with a recent bet on it pop up to the top of the list, rather than just the questions with the total # of traders (or other proxies for liquidity) staying consistently near the top?

Isn't that sorting by options where liquidity was recently added, because someone recently bet on it? I thought that was the system. In theory the highest liquidity questions shoudl be the ones with the most traders, which we can't sort by right now.

@Ziddletwix Yep. "trending" on the normal search page is totally different and unrelated to "trending" on a multiple choice market

@Gabrielle I dont' know enough to bet against you here but in the meeting today they were just talking about adding sort by liquidity as if it didn't exist already. So this probably resolves yes because I think they will do that but I don't think it's already true.

For anyone looking at this, discussion moved onto Discord. To summarize my point, https://discord.com/channels/915138780216823849/918236290439335936/1175243054081380493 shows @JamesGrugett saying

...Trending, which is just the amount of liquidity in the answer, which is typically based on the number of unique traders (b/c manifold injects liquidity for each new trader)

There were some questions about why that isn't the same as the number of traders, and I'm guessing that it's because market subsidies get thrown away when the market isn't at 50%. This is because on multiple choice markets p always equals 0.5. (For context on what that means, see https://manifoldmarkets.notion.site/Maniswap-ce406e1e897d417cbd491071ea8a0c39). This is according to James as well (https://discord.com/channels/915138780216823849/959728429295616061/1204508296703246396)

Yep, independent multi has p=0.5