4. Solution Strategy

A summary and explanation of the fundamental decisions and solution strategies that shape system architecture.

The DASTAN prediction market is a decentralized prediction market that allows users to create and trade predictions. Built on the Arbitrum One Layer 2 network, the prediction market uses an automated market maker (AMM) to price assets and an oracle to resolve markets. The prediction market is powered by a custom ERC-20 token called $POINTS, which is used as collateral for the prediction market. thirdweb tools play a critical role in the prediction market’s on-chain components.

4.1. Collateral

The collateral used for an on-chain prediction market is an ERC-20 token. We are using a modified ERC-20 referred to as $POINTS. $POINTS meets all ERC-20 token standards.

ERC-20 Modifications and other notes

  • $POINTS token is not transferable by the holder.

  • $POINTS token is transferable by select DASTAN addresses, to facilitate prediction market usage.

  • The addresses that are allowed to transfer $POINTS do not require approval to do so.

  • $POINTS has no max supply.

  • $POINTS are minted by user’s embedded wallets using a signature protected mint function.

4.2. Oracle

The prediction market’s oracle is essentially an address that is trusted to resolve markets. The address can be an Externally Owned Account (EOA) or a contract. The oracle is an account owned by DASTAN, but will be replaced by a protocol in future releases.

4.3. Automated Market Maker vs Order Book

An automated market maker (AMM) is a decentralized exchange that uses a mathematical formula to price assets. An order book is a centralized exchange that matches buyers and sellers based on their orders. While both models can be used in a decentralized prediction market, the DASTAN prediction market uses an AMM for the following reasons:

AMM Advantages

  • AMMs don’t require a central authority to match orders.

  • AMMs work well in low liquidity environments because they don’t require a counterparty to match orders.

  • The research and development going in to AMMs at this time is significant, and the DASTAN prediction market can leverage this work.

Learn more about AMMs from Gnosis.

4.4. AMM Algorithm

The AMM algorithm is a mathematical formula that prices assets. There are a multitude of existing AMM algorithms, many of which were created specifically for prediction market applications. The most popular algorithms are the Logarithmic Market Scoring Rule (LMSR) and the Constant Product Market Maker (CPMM). While each of these algorithms has its own advantages and disadvantages, the DASTAN prediction market will use the LMSR model for the following reasons:

LMSR Advantages

  • LMSR is used by many well-known prediction markets already, making it proven in the field.

  • LMSR is widely recognized in academic work, and it’s properties are more studied than CPMM.

  • CPMM is unable to account for markets with multiple outcomes, and is limited to working for a single outcome at a time.