Patent-Pending Framework

Seven forensic layers.
One verdict.

Every morning before the bell, Market Prism runs the same systematic analysis a quant desk runs manually — across every ticker in the coverage universe. Here's how each layer works.

Patent Pending: US 63/971,470 & 63/971,478
01
Narrative State Classification
Story lifecycle engine

Every market narrative follows a lifecycle: birth, growth, maturity, decay. Layer 1 classifies where each story sits in this arc using natural language analysis, attention velocity, and repetition entropy. A story that's been retold 40 times in 72 hours is structurally different from a fresh insight — even if the headline reads the same.

Breaking Surging Maturing Fragile Collapsing Stable
02
Narrative Decay Physics
Half-life modelling

Stories don't die linearly — they decay like radioactive isotopes. Layer 2 calculates the narrative half-life for each active story, measuring how quickly attention and conviction are dissipating. A story with a 12-day half-life that's 30 days old is structurally exhausted, regardless of how many analysts are still citing it.

Half-life estimate Energy remaining % Exhaustion predictor
03
SEC Filing Forensics
Text-vs-table divergence

Management teams tell stories in their MD&A sections — and reveal reality in their tables. Layer 3 measures the divergence between narrative claims and quantitative disclosures within SEC filings. High divergence is a leading indicator of earnings surprises, restatements, and narrative traps.

Prose sentiment Table reality Divergence score Credibility index
04
Coordination Detection
Herding analysis

When multiple independent sources converge on the same thesis within a narrow window, it looks like confirmation but often signals fragility. Layer 4 detects coordinated narrative clusters — when analysts, media, and social channels align too perfectly, creating a single point of failure for the prevailing story.

Source clustering Timing analysis Independence score
05
Bayesian Analyst Reliability
Track record weighting

Not all analyst calls are created equal. Layer 5 maintains a Bayesian reliability model for every source in our coverage universe, updating conviction weights based on historical accuracy, sector expertise, and timing precision. A bullish call from a source with 73% hit rate on semiconductors carries different weight than one from a generalist with 41%.

Source reliability Sector accuracy Historical calibration
06
Fair Value Stress Testing
Price-narrative gap

Layer 6 compares where price is against where the narrative says it should be. Using a multi-factor model that combines fundamental anchors (P/E, FCF yield, growth rate) with narrative energy, it identifies tickers where the gap between story and reality has become structurally unsustainable.

Fair value estimate Gap magnitude Sustainability score
07
Contagion Mapping
Cross-ticker propagation

Stories don't stay contained. Layer 7 maps how narrative shocks propagate through supply chains, sector peers, and thematic clusters. When NVDA's China export story shifts, Layer 7 traces the impact upstream (TSM, AMAT, ASML) and downstream (SMCI, data center REITs) — before the market prices it in.

Upstream tickers Downstream tickers Propagation speed Contagion risk
Final output

Three verdicts. Zero ambiguity.

All seven layers converge into one of three actionable verdicts — not a score, not a probability, but a clear structural assessment.

Narrative Trap
Price is being driven by a story that the forensic evidence doesn't support. The gap between narrative energy and structural reality is widening.
Structurally Supported
The underlying narrative aligns with fundamentals, filings, and sector dynamics. The story is credible and the price action is justified.
Monitoring
Evidence is mixed or evolving. The signal isn't strong enough for a conviction call — the system is watching for a structural tilt.