Author Topic: Paper "On Decentralizing Prediction Markets and Order Books"  (Read 2041 times)

0 Members and 1 Guest are viewing this topic.

Offline Fairlay

  • Newbie
  • *
  • Posts: 6
    • View Profile

Much of the paper was devoted to how to organize a order book for prediction market trading. Truthcoin, takes a much more effective approach using Robin Hanson's Market Scoring Rule, which seems to be getting overlooked by centralized prediction markets like Predictious and Fairlay. The Market Scoring Rule can allow for instant buys and sells.  Order books would require you to wait for someone else to take the opposing side.


Obv. we have a problem with liquidity until now - we will give market scoring rules (or a modification of it) a try!

Offline Bitcoinfan

  • Sr. Member
  • ****
  • Posts: 240
    • View Profile
past due commentary on this:

Much of the paper was devoted to how to organize a order book for prediction market trading. Truthcoin, takes a much more effective approach using Robin Hanson's Market Scoring Rule, which seems to be getting overlooked by centralized prediction markets like Predictious and Fairlay. The Market Scoring Rule can allow for instant buys and sells.  Order books would require you to wait for someone else to take the opposing side.

Voters as shareholders in a company.
Essentially a public company that votes on the outcome of a market.  Shareholders vote on the outcomes of prediction markets based on their proportional to their ownership of the company.  The paper makes a case that voters are disincentivized from misbehaving because a false vote would result in the bankruptcy of the company and therefore future earnings. 

This is very similar to Paul's idea.  However I think Truthcoin is much better because it puts constraints on voters through SVD.  So whereas in this paper someone with a large stake can swing the consensus and pull everyone to his position, SVD mathematically weighs the consensus across all the votecoins for that period.  Even with a 50%/50% tie, SVD is able to find the consensus because the Truthcoin voters who voted Yes matched more closely together.

Voting as a Keynesian beauty contest.

Sounds like a arbitration formed based on having a pool of voters who post a collateral bond.  At the end of an outcome, votes are tallied, where a majority decision is called when 2/3 of the voters agree.  The 1/3 left behind loose their collateral. 

The paper acknowledges that there is a risk.
" If either coalition is able to convince the other that they are absolutely going to spend their N/2 votes on their preferred outcome, the other side is incentivized to back down and concede to prevent losing their bond."
 
I think is probable, since due to the transparent nature of posting a large bond publicly opens doors for collusion.  People would therefore have many communication channels to reach a voter and convince them that the group is voting against the right outcome. 

Paul's Truthcoin is great because voters are distributed into many smaller shareholders, where it would be impossible to get in contact with everyone before a prediction market is decided. 

Offline Bitcoinfan

  • Sr. Member
  • ****
  • Posts: 240
    • View Profile
Quoted from the whitepaper: "Our decentralized PM is strictly a design. We are not currently pursuing any level of actual deployment."

Otherwise I have a bit to say about this.  I'll get to it tomorrow.

Offline toast

  • Hero Member
  • *****
  • Posts: 4001
    • View Profile
  • BitShares: nikolai
I skimmed it and it seemed like nothing more than a description of PMs and how to implement them on a blockchain - though a very thorough and well-written one. They do not have insights into how to stop miner front-running (our order matching algo makes it irrelevant) and they don't allow themselves much flexibility in implementation because they assume you'd be building this on bitcoin/coloredcoins
Do not use this post as information for making any important decisions. The only agreements I ever make are informal and non-binding. Take the same precautions as when dealing with a compromised account, scammer, sockpuppet, etc.

Offline sudo

  • Hero Member
  • *****
  • Posts: 2255
    • View Profile
  • BitShares: ags

Offline santaclause102

  • Hero Member
  • *****
  • Posts: 2486
    • View Profile
I don't have the time to read it atm. But it might be of interest for BTS X and in general for some of you.

http://users.encs.concordia.ca/~clark/papers/2014_weis.pdf