Skip to main content
MANIFOLD
Would boosts be better if we were paying people to trade in a market rather than to just click a button?
13
Ṁ270Ṁ288
resolved Jun 17
Resolved as
59%

Resolves according to the market PROB.

I feel like it would but I might be missing something.

Market context
Get
Ṁ1,000
to start trading!

🏅 Top traders

#TraderTotal profit
1Ṁ6
2Ṁ4
3Ṁ2
4Ṁ1
5Ṁ1
Sort by:

Resolves according to the market's choice.

is this resolves PROB to market or resolves to poll?

Anyway people who do not have any useful information will just randomly trade, so it'll reduce accuracy, not improve it

@jacksonpolack Prob, thanks. Updated

Prediction market accuracy would probably suffer if people are incentivized to trade on markets they don’t care about.

There’s a reason the whole real world ad marketplace works on views/ad clicks and not “buying the product advertised”, too. Ultimately the “product” should need to be interesting and high quality to benefit from ads.

predictedNO

It's also worth noting that markets that got less clicks per view would stay on people's feeds longer before exhausting their budget.

That means on average people get crappier, less relevant feeds, as bad markets functionally get more views per Ṁ invested.

@DanMan314 Nice, this is a great point.

@DanMan314 That’s not true - plenty of advertising campaigns work on a “cost per acquisition” basis.

that said in this case an uninformed trade has no value to a market owner, while an arbitrary acquisition has value to the company selling the product.

(or an uninformed trade should have no value, maybe some people want trader bonuses and other people want their social relevance number to go up)

People still have an incentive to bet in the direction they think is more likely, right? It looks to me like at best, we get some wisdom of the crowds effect. At worst, it’s injecting some dumb money into the market for increased liquidity for knowledgeable traders to take.

People still have an incentive to bet in the direction they think is more likely, right

There are a lot of markets - most markets, actually - that I don't bet on because I have no idea what the right answer is, no valuable information whatsoever.

For instance:

Will John Riccitiello be fired, forced to resign, or otherwise announce termination from Unity Technologies by end 2024?

huh?

It looks to me like at best, we get some wisdom of the crowds effect

The wisdom of the crowds effect isn't useful if the crowd doesn't know what the question is about. If you poll a thousand people on "should Rust use Return Type Notation or Higher-Ranked Projections", you are not going to get anything useful. Because none of the people you're polling know what those are!

At worst, it’s injecting some dumb money into the market for increased liquidity for knowledgeable traders to take.

There's a reason manifold markets don't start with M$1k liquidity, small traders need to be able to move the market if small traders are all the market gets.

One downside is that they could still buy and sell M3 worth of shares immediately, making more noise than anything else productive. Would also then have a net inflationary effect rather than deflationary, as the unique trader bonus would create M5 almost every time.

@brubsby I think it’d be fair to remove the trader bonus for subsidized traders.

Good point about the noise. Maybe you lose the bonus if you sell your position within a week after trading.

predictedYES

@IC I think means testing the bonus too much is inelegant, but I do like the idea of giving people money that they are forced to bet with. Maybe the idea of "locked" shares that become open after a while, perhaps implemented by being owned by manifold and then paid out when they "vest"