The ThinkMarkets group combines deep experience and almost a decade’s-worth of data acquired on how tens of thousands of traders react and interface on our trading platform. This knowledge allows us to create complex scoring models that will tag each trader on the network with various scores. These scores will be embedded into the smart contract off-chain to allow makers and takers to match each other better.
The units of this scoring will be called ‘Personas.’ 

Personas are algorithms that will ‘learn’ how a Trader interacts on the network and build a roadmap so each trade is matched on the network with the right participant as quickly as possible and at the best price. As the TradeConnect network grows, personas will be become smarter and improve in real time as trading volume grows and there is a larger pool of subscriber data to build from. 

Trade Quality Persona 

This is the score that will model how a specific trade interacts with a given financial instrument over the course of the lifetime of the trade and the subscriber placing it. For example, by modelling when a given trade was opened, closed, changed for its stop loss, the level of profit and how the trade is managed by the subscriber we can measure the ‘quality’ of a trade against other participants in a pool. The score will adjust higher or lower based on the quality of the trade and will be tagged with the Trade Quality Persona. 

As the volume of trades on the network grows the vast quantity of data will allow the trade quality persona to serve as a reference point for maker/takers in their bidding for the trade. For example, a taker can choose to only accept trades that have a high-quality score based on the open times of the trade, and thus is willing to reduce the spread and commissions for the trade. 

Price-matching Persona 

Participants in the network will want to capture trades from traders that earn them the highest yield in spread and the lowest risk. For any given trader as trades are matched peer-to-peer or any of the various other matching options on the network,

The price matching persona will create a score to capture the prices that a participant trades in relation to the rest of the participants in a pool. The score will adjust higher or lower depending on how close to the ‘mid-point’ or better or worse it is in relation to the rest of the network. Each trade on the network will be tagged with this Price Matching Persona and will encourage makers and takers to bid for the right to match the trade. As participants on the network grow, this bidding will create the best pricing scenario for each trade.

Did this answer your question?