Skip to main content

Chapter 7: The Vector Engine: Market Trend & Momentum Analysis

7.0 Abstract​

Beyond new assets, the Vector Engine is designed to identify and predict momentum within the broader market of established tokens. It moves beyond simplistic metrics like 24-hour volume to create a holistic, real-time "Momentum Score" for thousands of assets, allowing METAsol to position capital before trends become obvious to the general public.

7.1 Calculating the Momentum Score​

The Momentum Score is a proprietary, weighted score from 1-100 calculated for every liquid asset on Solana. It is a composite of several key data factors:

  • Volume Velocity (30% Weighting): The rate of change in trading volume. A token whose volume is accelerating rapidly scores higher than one with high but stagnant volume.

  • Trade Delta (25% Weighting): The net difference between aggressive market buys and market sells. A high positive delta indicates strong buying pressure.

  • Wallet Concentration Flow (20% Weighting): We track whether a token's holder base is becoming more concentrated (often a bearish sign of whales accumulating before a dump) or more distributed (a bullish sign of retail adoption).

  • Prime Intelligence Inflow (15% Weighting): A significant factor is whether wallets categorized by our Prime Intelligence module as "profitable traders" are accumulating the asset.

  • Social Sentiment (10% Weighting): We ingest and analyze real-time data from platforms like X (Twitter) and Telegram to gauge social media hype and sentiment around a token.

7.2 Identifying "Hot Pairs" and Capital Flow Concentrations​

The Vector Engine aggregates Momentum Scores to identify "Hot Pairs" and sector-wide trends. By analyzing the flow of liquidity out of one pool and into another, we can predict capital rotation.

For example, the engine might detect that liquidity is flowing out of older meme coin pools and into new GameFi token pools. This insight allows the broader METAsol engine to preposition capital to benefit from the emerging sector-wide trend before it becomes saturated.

The historical data generated by the Vector Engine is used to train machine learning models that can predict trends with a high degree of accuracy. These models can identify complex, non-linear correlations that are invisible to human analysts, such as the relationship between NFT market activity and the subsequent performance of a specific DeFi token. These predictive insights provide the highest level of strategic direction for our capital allocation.