TL;DR: Resolves 80% to the market I put 10K subsidy into, and 20% to the market I put 1K subsidy into. N/A if I do not put a 10K subsidy into any market.
As I've been subsidizing markets for @Austin's Public Interest Subsidy Program, I find myself coming to the conclusion that bigger is better. The existing systems of Manifold already do a decent job of providing subsidy to small questions. When I bring a question from M100 liquidity to M1100 liquidity, I'm not sure I'm doing something that Manifold wouldn't have already done with house subsidies.
But until the recent subsidy changes, this house subsidy was capped at M2000 liquidity for Yes/No questions. And even after the recent update, adding just 5 mana per trader up to 10,000 traders hasn't added much to any of these questions yet.
So I think we should do it ourselves! These are important questions with thousands of traders, but the markets can be moved with just a few dozen mana! These markets should all be much thicker, in my opinion.
Many of the best traders of the site have few positions on these markets. Now, that's understandable given that high-trader markets are necessarily about difficult questions that don't resolve for some time. There's less incentive to trade on such questions, when you can trade on questions you're more confident about and which resolve sooner.
I think this is a valuable chance to experiment with putting large subsidies into already-popular questions, to see if it can attract a new wave of interest and increase the market's accuracy.
Accordingly, I am planning on putting M10,000 of subsidy into one of the questions I've listed above. I will also put an M1000 subsidy into a second of these questions, and resolve the market 80%/20% between those two questions.
I have a favorite in mind, but I'm not going to say what it is at first because I'd like to see if the market agrees with me. If you have other suggestions besides these markets, post them in the comments and I will add them as an option if I think they're worth considering.
If I do not put an M10,000 subsidy into any market in the next week, this market will resolve N/A.
I will not trade on this question beyond buying down Other initially and then selling my no shares if a new question is added, and I will provide additional information about my decision in the comments as time passes.
1,000
3.00The market has succeeded in deciphering my intent! My plan is to put the 10K into /LarsDoucet/will-joe-biden-win-the-2024-us-pres. Let me lay out my case for this, to see if anyone can talk me out of it.
First of all, the market has 2.4K liquidity. The market on Biden's nomination has twice that, and the market on the speaker of the house has 25 times that! I think we should have a lot more liquidity on what is probably the single most-predicted question in all of forecasting right now.
Second, look at who has large positions!
I follow a lot of people, especially active members of the community. Usually on a big market, I'll see a lot of highlighted names on both sides of the positions. But here, it seems almost no one is willing to bet big on Biden! This is espeically interesting because third:
Electionbettingodds has the average price of Biden in real-money markets at 33%! That's such a large difference! I'm very curious to see if a large influx of neutral mana into the market moves it closer or further away from the real-money markets. And I intend to find out!
The 1K subsidy/20% resolution is still up for grabs, and I might increase it from 1K since I think that might not make a big difference to most of these markets.
But I'm pretty sure I want to give the 10k/80% resolution to the general election market.
Betting 10k on YES moves the probability up 3% to 59, and betton 10k on NO moves the probability down 4% to 52. That's a lot more resistance than an extra 10k in subsidies will bring. This is the kind of market that'd almost be liquid enough without an AMM tbh
20k moves it down to 50 or up to 60.
The limit orders are good of course, but I think the market should still have more neutral mana even between those orders. I don't expect anyone to come in and suddenly move the market to 30% or 70% after I add liquidity, but I expect it to increase activity. A big market like this should be able to take hundreds of mana of bets within the same percentage point, IMO.
A big market like this should be able to take hundreds of mana of bets within the same percentage point, IMO.
If you think that's worth 10k mana go ahead, but I don't think there's anything else. A 20% increase in the amount of mana you can win on the market won't increase interest to an appreciable extent imo.
If you're correct that's probably true regardless of which of these markets I chose, so I recommend buying some no in This Market! I don't think any of this has talked me out of trying it, though I appreciate the feedback of course.
Got some more good thoughts from Discord, but still nothing that's talked me out of it so far. So now all that remains is for me to pick one of these other markets to give the 1K subsidy to. The market still has the AI Olympiad market higher than the rest of the political markets and it was ahead from the start, so I think I'm leaning towards that one for the 1k!
There is a much higher than normal probability of there being no presidential debates this cycle. Especially if Trump continues to behave in an unhinged way. Who cares if he is the GOP nominee, they are just a fringe party now anyway. Won't hurt Biden at all to dismiss the fringe lunatic in the Fulton County prison chosen by the insurrectionist party to be their nominee for president. I have said it before and I will say it again, Trump will barely outperform Barry Goldwater. 90 million people will vote against him and he will lose the popular vote by 15mm (+/- 2.5mm). He will only win half of the states he is supposed to and even then it will be by the narrowest of margins because in 2024 every state is going to be Georgia. So don't waste your subsidy on a market that might N/A.
@RobertCousineau Per the comments below it's still not going to be that one, unless you can talk me into it!
@lag in hoping for the ai olympiad one, cause it's more interesting than the political markets and I have yes shares on it
@lag this is just a how manifold works question, but if i bet an option from above 20 to 39 and and the option was later brought down to lower than 20% and was resolves 20%, do I get mana from my yes shares, as i wouldve bought shares above the resolution point but the option was below 20% at resolution
@lag this is just theoretical, as i bought shares from 19 to 39, but it would still sorta apply to me in the current state of the market
@lag yeah the percentage at time of resolution doesn't change the value of your shares. Every yes share would pay out for .2 mana instead of 1 mana.
@lag as to your point about the nomination markets, I agree that because they are both very likely to resolve Yes that they aren't the very best candidates for the 10k subsidy on the list IMO.
That said, I'd still subsidize all of these markets with 10k if I could. But this is about picking the one I think is the best test case.
I'd like to suggest https://manifold.markets/IsaacKing/did-covid19-come-from-a-laboratory. I think it's by far of the highest public interest of the large questions. However, I'm slightly worried that this market is unresolvable. I don't know if any more information will ever come out.
How about the Sudoku market? /Mira/will-a-prompt-that-enables-gpt4-to
The M25k subsidy is only matching people's API costs.
I think it would be a good idea to help cover some of those costs more! It brings a good amount of attention the site. But I don't know that more liquidity will help much, Emily actually turned me down for a loan that would let her buy more Yes.
I think a straight up bounty prize or literal cash prize for the person to get an 80% solution would be a good investment on Manifold's part though.