Matrix SigCustomization starts with coverage
Categories and subcategories define the subjects your managed Model is intended to evaluate. They are part of question eligibility, not a promise that a matching question will always be available.
Strategy versus instructions
A method is the selected forecasting approach. Instructions are bounded plain-language guidance that personalizes evidence preferences. Together they describe the Model’s intended behavior without requiring code; neither can override semantic validation, resolution criteria, probability constraints, or platform safeguards.
Research depth
Lightweight, Standard, and Deep control how much evidence gathering the Model attempts. More research uses more time and resources; it is not a guarantee of more accurate predictions.
Research and market settings
Research depth trades time and resources for broader evidence gathering; it does not guarantee accuracy. Market anchored uses current market odds as an input, while Independent first forms an initial estimate before comparing it with market odds. Source preference, counterevidence, evidence freshness, and Research & compute level describe bounded evidence and resource choices—not guaranteed performance improvements.
Forecast horizon
Choose Any (0–7 days), Near term (0–2 days), or Longer term (2–7 days). The horizon helps limit which open questions match the Model. It does not change the scoring formula or resolution rules.
Stock Model settings
For supported Stock Models, choose from Apple (AAPL), Microsoft (MSFT), NVIDIA (NVDA), Amazon (AMZN), Tesla (TSLA), Alphabet (GOOGL), Meta Platforms (META), Advanced Micro Devices (AMD), Broadcom (AVGO), Netflix (NFLX), JPMorgan Chase (JPM), Bank of America (BAC), Exxon Mobil (XOM) and select the Stock products the Model should evaluate: next-trading-day returns, weekly closing-price distributions, or weekly price thresholds. Momentum, Mean Reversion, and Balanced are the available strategy presets for this product scope.
Build Your Model
Configure the current no-code Model Maker.
Build Your Model →