Scoring Engine How conviction, AI resilience, ownership intelligence, and financial health scores are computed

Platform Overview →

The pipeline uses a multi-layered scoring framework to evaluate investment opportunities. Every score is deterministic (reproducible from the same inputs), except AI Resilience which uses LLM-generated assessments calibrated against reference anchors.

How to read this guide

Each module page shows the components, formulas, thresholds, and data sources. Scores are 0–100 unless noted otherwise. All thresholds and weights are configurable via the score_config table.

Core Modules

Conviction Score

The primary investment signal. Four weighted dimensions: Fundamental, Thematic, Valuation, Catalyst. 0–100 composite with label thresholds. Catalyst now includes insider activity and institutional momentum.

AI Resilience

Five-dimension LLM-scored framework assessing how a company is positioned for AI disruption. Revenue Catalyst, Moat, OpLev, Pricing, Obsolescence.

Financial Strength

Balance sheet snapshot + three YoY trend signals. Sector-aware industry adjustments. 8 component ratios from quarterly financial data.

Signals & Trends

Estimate revisions, PE momentum, theme momentum, rating momentum, peer rankings, thesis consistency, insider activity, institutional momentum, price path flags. TTL-based unified signal store.

Ownership Intelligence

SEC Form 4 insider transactions with relative scoring (% of holdings, baseline comparison). WhaleWisdom 13F institutional holdings with accumulation/distribution signals and HHI concentration.

Price Path Analysis

Index-level and company-level technical analysis. PE decomposition (multiple expansion vs earnings growth), RSI, Bollinger Bands, stretched score. Target accuracy tracking.

Architecture Overview

conviction_scores (composite output)
  |
  +-- Fundamental (25%)
  |     beat_rate, surprise_quality, sentiment,
  |     estimate_trajectory, guidance, financial_strength
  |
  +-- Thematic (25%)
  |     ai_resilience (adaptive weight), theme_health, peer_standing
  |
  +-- Valuation (25%)
  |     pe_percentile, peg, price_strength,
  |     net_debt_ocf (TTM, sector-aware),
  |     ev_fcf (TTM, sector-aware), pe_momentum
  |
  +-- Catalyst (25%)
        thesis_conviction (calibrated + penalty),
        rating_momentum, estimate_momentum,
        guidance_trajectory, thesis_consistency,
        insider_activity, institutional_momentum,
        contrarian_signal

Data Flow

Raw data flows through three layers before reaching conviction scores:

Layer	Source	Output
Ingestion	FMP, SEC EDGAR (XBRL, Form 4, 8-K, 10-Q/10-K), FRED, WhaleWisdom, CarbonArc, Reddit, newsletters	earnings_events, financial_health, companies, news_articles, analyst_ratings, insider_transactions, institutional_holders, filing_extracts
Signals	Ingested tables + LLM analysis	ticker_signals, ai_scores, theme_momentum, peer_rankings, estimate_revisions, insider_signal, institutional_signal, price_path_flags, target_accuracy_snapshots
Scoring	Signals + company data	conviction_scores, investment_theses, earnings_previews, decision_scorecards

Custom Sector Taxonomy

The platform uses a custom sector taxonomy (not GICS). Technology is split into four groups:

Sector Group	Sub-sectors
Semiconductors	Logic/GPU, Networking, Analog, Equipment, Memory, Foundry
Software	Enterprise SaaS, DevOps/Data, Cybersecurity, Digital Advertising, EDA, Fintech, Mobility, Gaming
AI Infrastructure	GPU Cloud, Quantum
Cloud/Hyperscalers	Cloud/Hyperscalers
Energy/Power	Utilities, Renewables, Nuclear, Midstream
Industrials	Defense, Infrastructure, Enterprise IT
Healthcare	Pharma

All peer rankings, sector reports, and dashboard groupings resolve through this taxonomy via SECTOR_GROUPS (Python) and _peerKey() (JavaScript).

Configuration

All weights, thresholds, and labels are tunable at runtime via the score_config table without code changes:

Config Type	Key Settings
conviction	Dimension weights (default 25% each), label thresholds (Strong Buy ≥80, Buy ≥65, Hold ≥50), macro modifier toggle
ai_resilience	Dimension weights (RevCat 25%, Moat 25%, OpLev 15%, Pricing 20%, Obsol 15%), label thresholds
financial_strength	Industry adjustment factors (Energy 1.15, Utilities 1.15, Industrials 1.10)
risk_tiers	Anchor ≥70, Ballast 60–69, Contrarian 50–59, High-Risk <50