How It Works

PatternRank turns historical market behavior into model-derived equity rankings. Each workflow is designed for repeatable research review, measured outcomes, and clear limitations.

1. Data Collection

Historical market data is collected for liquid US equities traded on NYSE and NASDAQ. The pipeline emphasizes consistent inputs across price, volume, liquidity, and post-ranking outcome data.

2. Pattern Recognition

Proprietary quantitative models analyze historical setups, momentum shifts, and structural filters. The goal is a ranked research queue with confidence context, not a recommendation list.

3. Ranking Generation

Models produce ordered rankings based on confidence scores and available market context. Each ranking is presented with supporting metrics for independent evaluation.

4. Daily Updates

Rankings are refreshed after market close. Filter by model type, date, confidence, and coverage universe to keep the review process focused.

Ranking Models

7D Momentum Rankings

Short-term momentum behavior ranked over a 7-day review horizon using proprietary quantitative pattern analysis and measured follow-through.
  • 7-day research horizon
  • Model confidence context
  • Post-ranking outcome tracking
  • Confidence-ranked results

ETB High-Quality Rankings

Liquidity-filtered rankings focused on easy-to-borrow US equities commonly tracked by professional desks. Designed to reduce structural noise in the review universe.
  • Institutional-grade liquidity filtering
  • Easy-to-borrow universe
  • Quality-focused pattern rankings
  • Reduced structural noise

Important: Educational Purpose Only

PatternRank is an educational analytics platform. PatternRank is an educational analytics platform. Rankings are for research and informational use only and are not investment advice. Past performance does not guarantee future results. Users are solely responsible for their own decisions. Please review the full disclaimer before using the platform.

Ready to review the ranking workflow?

Start with sample rankings, then choose the level of model coverage that fits your research process.