$NTDO | $PMRT markets live on beta now
For the complete documentation index, see llms.txt. This page is also available as Markdown.

Oracle design

Oracle price combines multiple price sources to try form the most un-bias market price.

Oracle price is formed from summing the total of mark (last traded price) + validator (25+ data sources, ie, collector, courtyard, phygitals) + EMA and dividing by 3. Each data point takes up ⅓ of weight.

As more volume and price data is formed on Trove to create a liquid market for live pricing, we can see certain weights (ie mark or EMA) take up more pricing weight (>33%). Pricing weight will be tested and vary in the Trove beta.

Oracle composition

Design Principles

External data grounds the oracle in real-world pricing. Internal signals ensure responsiveness to live market sentiment.

Auction houses, marketplace sales, floor listings, and verified price books from 25+ sources like Courtyard, Collector_Crypt, Phygitals, and TCGPlayer.

Live mark price, executed trades, order book depth, and open interest metrics from Trove’s on-chain markets.


How the Oracle Is Calculated

Mark: $31.5 (last traded price on Trove) - The last traded price on Trove, reflecting real-time trader sentiment and market action.

Validator: $31 (sources sum/ quantity or sources) - Aggregated data from 25 verified external sources, including Collector_Crypt, Courtyard, Phygitals, TCGPlayer, and others.

EMA (HyperPS Model): $31.8 - Trove’s proprietary exponential moving average model, smoothing volatility and tracking short-term pricing trends.

Each input starts with ⅓ weight, ensuring that no single component dominates the oracle.

This tri-layered design anchors pricing in external reality while maintaining agility through internal market signals and predictive modeling.

Last updated