skininthegame

What are some examples of prediction markets getting it wrong? 
E.g. prediction markets believed Hillary would win the 2016 US presidential election. 

Trying to build a comprehensive list of examples. 

Also curious as to strong counter-examples I.e. prediction markets were right & contrary to the “professionals”

is it more correct to view an efficient prediction market as a mechanism to provide accurate probabilities for outcomes, rather than predicting binary outcomes? 

e.g. prediction markets correctly assigned Trump a <50% probability of winning the 2016 election.

dev, armchair economist & crypto tea leaves reader - 

/futarchy /network-states x.com/nicovrg_




And if so, what is the implication for conditional markets i.e. futarchy? @nicovrg

to me prediction markets have an evaluation "resolution" that reward people who were right whereas conditional market protect holders & allow them to express their desire depending of proposal result

example ->

Hmm. This makes sense but doesn’t answer the question I’m trying to articulate

Another way stating this: should we expect conditional markets to “get it wrong” some % of the time? If so, is there a correlation between the pass v. fail market prices and the probability the “worse” proposal passes

there might be some keynesian beauty contest in the sense that people trading conditional markets try to predict what other traders are going to trade. you can think that a proposal is -ev (should bid fail) and bid pass because you assume other traders will view it as +ev or do the same assumption as you

its not always (more often than not) possible to know what would have happened if the other market would have passed since you that reality didn't materialise. its also hard to isolate impact of different proposal on the token price months away from now. which is responsible for what?

futarchy is going from a thought experiment to fast iteration cycle in reality. we'll figure things out as we go - binary outcomes will change when we get multi modal proposals - and i already have ideas to get number out instead of yes/no or option[x]

all in all it remains an experiment and there is so much to do that i didn't even touched on here, which is why it's exciting and we're here 🔥

then we'll see how we do for future forecast. i'm not the one (no one is) that decide what will happen, but could see "evaluation" or "contracts" added to to help fix this potential issue