Will Questions of the form "Will this question go below X% before R" be banned by manifold markets before 2025?
➕
Plus
19
Ṁ1270
Dec 31
7%
chance

Assume we have "Will this question go below 20% before Resolution date" and you see the market currently at 25%. If you say yes, it bumps the market value up, away from the critical vlaue. If you say no, you move it towards the critical value of resolution, but against your own potential to profit at resolution,

An analogous but inverted situation happens for "Will this question not exceed 80% by resolution"

You observe the market at 75%, if you vote no, you increase the chance of a positive resolution, if you vote yes you decrease that chance.

Both situations are paradoxical, and despite their beautifly Russellian nature, It seems like a market makers scam.

This market will resolve by, on the 25th Dec 2024, my attempting to make the ADDITIONAL MARKET

"Will this question go below 10% before 00:00GMT 30th Dec 2024"

If that ADDITIONAL fails to resolve and is banned/blocked/removed/impossible to create,

THIS market will resolve YES

If it is not banned/blocked/removed/impossible to create, this market will resolve NO

Simple as <3

Get
Ṁ1,000
and
S1.00
Sort by:

Variants of questions like this can actually help address some of Manifolds other incentive issues. The superconductor market, if it resolves NO, won't resolve for a long time. If it resolves YES, it will resolve sooner. Despite the existence of loans, This still distorts incentives in favour betting YES because of this. Look at the superconductor market hat doesn't resolve until 2030 - it is trading way too high with little incentive to bid it down.

Derivative markets can help superconductor NO betters have something to bet on that resolves sooner.

It's true that I "manipulated" the superconductor market to get it to go below 30%, and immediately sniped the "will it go below 30" market. In the case of the superconductor market, this would have been a bad idea if I didn't think 30% was a good price to buy NO at. In fact, I mostly did this so that I could cash out of the "will it go below 30%" market sooner and free up mana to buy more NO on the superconductor.

For less liquid markets I'm not sure how much of a manipulation risk it is. Either there are no limit orders and @acc thinks the user is well-calibrated, and the user immediately reverses their trade. Or they have to go through @acc and/or some limit orders, which they should expect will lose them mana if this was not a good price given their beliefs. What's the harm done?

The actual underlying market is still bound to reality in the end, so manipulation is costly.

Though I have not much interest in markets about manipulation, so if it were me I'd try to make resolution criteria that, e.g. required.the probability to average below X for Y days or similar, just to make sure the market was measuring more signal than noise (or manipulation)

@chrisjbillington Ironically there is the same problem on this market. It doesn't resolve until 2025, which means people have little incentive to bet it lower. I think it should be lower than 10%.

@ShadowyZephyr 100%, the long-term market problem remains unsolved. I didn't quite appreciate it as much in the past and so still made lots of bets on longer-term markets, and I now think that the only reason long-term markets aren't even more mispriced than they are is because people bet despite the poor incentives, because they're not thinking too hard. The more profit-optimising traders are, the worse long-term markets will be mispriced.

I hope for some rule or incentive that avoids those questions

There are enough ways to do this kind of self-referential nonsense that they will have trouble designing a rule that they can apply consistently, and will be inclined to go back on the initial design if they attempt one.

@JubileeYoung You persuaded me to sell

predicts NO

@TamarSpoerri ahaha. there's plenty of reasons to buy "yes" imo, it's just that I've seen the history of the trouble of moderation. there's still room for "they walk it back, then walk back the walking back", but then there's also room for "and then they walk back the..." oh nooo recursive self-reference STRIKES AGAIN!!!

Feel free to ask for any clarifications if ive left gaps in my explanation.