Seven dimensions. One multiplicative formula. No black boxes. If any dimension scores zero, the whole thesis dies. That's the point.
Why multiplication, not addition? Because a fatal flaw is fatal. A zero in any dimension — untested management, zero catalyst proximity, a thesis the Delbert test can't explain — collapses the whole score to zero. Addition lets you hide a fatal flaw behind high scores elsewhere. The Sextant doesn't allow that.
Each dimension is scored 1–10. The sub-scores within each dimension are auto-computed or manually set by JD, then combined into the dimension score. The composite formula above normalizes the result to a 100-point scale called the Gray Street Score (GSS). No dimension score ever returns zero or null. Scale is 1–10 only. Floor of 1 enforced everywhere in the pipeline.
The market doesn't understand the company. Analyst coverage is thin. Short interest is elevated because the story is counterintuitive. The 52-week position reflects prolonged neglect. The gap between the price target and current price is wide. All four of these create the setup — not any one alone.
The Delbert version: Is the stock cheap because nobody's looked at it, or cheap because it deserves to be cheap? Misunderstanding is the first check.
Something is coming. Volume is spiking above normal. RSI is in oversold territory. SEC 8-K filings are accelerating. These are the pressure gauges. A good thesis with no catalyst is a boat with no engine. C measures how close the detonation point is.
Is this company doing something that threatens an established market structure? Not incremental improvement — structural displacement. D is set manually because the kind of disruption that matters requires qualitative judgment. The scanner cannot read a patent filing and understand that it rewrites the supply chain physics of an entire industry.
The Delbert version: Could this company put somebody big out of business, or just take some market share?
The asset base, technology, or IP is vastly larger than the company's market cap implies. Nobody has noticed. A gold mine priced like a mailbox. A patent portfolio that controls the next generation of a market sitting on the balance sheet at cost. S measures how deeply the market is undervaluing what's already there.
The Delbert version: Is there a lot more there than the price tag suggests?
The company controls a critical input, process, or geography that others cannot easily replicate or substitute. Chokepoint value is structural — it persists through cycles. CP is elevated automatically for companies in sectors the framework classifies as strategically critical.
CP is self-reinforcing when CP reaches threshold in a strategically critical sector: the framework treats this as an additional resonance condition (see Resonance, below).
The Delbert version: If this company disappeared tomorrow, who would have a serious problem?
Can the team execute? Mg-Q uses insider ownership percentage as the primary proxy. Skin in the game is the most honest alignment signal available from public data. Founders who own 20% of the company behave differently from hired executives who own nothing.
Has management been honest? Mg-T cannot be automated. It requires reading earnings calls, tracking guidance accuracy, and applying judgment built over years of watching people talk about money under pressure. Default is 7. Adjusted manually, almost always down, never without a documented reason. Low trust disqualifies the thesis even if every other dimension is screaming.
The Delbert version: Would you lend these guys your truck?
Analyst coverage, short interest level, 52-week price position, gap between price targets and current price.
Volume vs recent average, RSI technical position, SEC filing velocity.
Market size relative to company valuation, revenue trajectory vs sector peers, structural position in critical supply chains.
Analyst coverage depth, recency of institutional attention, 52-week price position, ownership trend.
Sector classification, structural control of critical input or process.
Insider ownership percentage.
Manually assessed — earnings call accuracy, guidance history, management behavior under pressure.
The GSS is the foundation. Resonance is a separate layer built on top of it. Resonance scores pattern combinations — when multiple dimensions align in ways that historically precede large moves. Hard cap applies when management trust falls below threshold.
Superset suppression: Tidal Wave (S+M+C) absorbs both the standalone M+C pattern and the C→M pattern when all three co-occur. You don't double-count them.
S + M + C all elevated simultaneously. The asset is deeply undervalued (S), widely misunderstood (M), and a catalyst is forming (C). This is the complete setup. When all three arrive together the move, when it comes, tends to be large and fast.
Superset: absorbs the Underdog Rally (M+C) and the Pressure Cooker (C→M) when active.
M + C elevated. The stock is misunderstood and a catalyst is forming. Standalone pattern — significant but smaller than Tidal Wave. Absorbed by Tidal Wave when all three (S+M+C) are present.
S + D elevated. A sleeping giant with disruptive technology or structural position. Only evaluated when D has been scored for the ticker. The combination of "more there than priced" and "threatens an established order" is a particularly dangerous combination for short sellers.
When CP reaches threshold and the sector is classified as strategically critical: structural control of a critical resource or process doesn't just protect margin — it compounds. Each quarter at the chokepoint is a quarter competitors can't replicate.
High catalyst proximity meeting persistent misunderstanding. The market is wrong and the evidence is building. This is not a subtle setup — the disconnect between fundamentals and price is measurable and the resolution is closer than the price implies. Absorbed by Tidal Wave when S also elevated.
M + CP + D all elevated. Misunderstood company with a structural chokepoint and disruptive technology or position. Rare. When it appears the combination of misunderstanding, structural moat, and disruptive upside creates a compounding asymmetry that the market almost never prices correctly.
All seven dimensions elevated simultaneously. The complete Sextant signal. Every dimension of the thesis is firing at once. In three years of running the framework across several hundred tickers, this has occurred fewer than a dozen times. When it does, it gets immediate priority research attention.
Empirically discovered, March 2026. Volume or technical pressure at elevated levels. This is not a thesis pattern — it's a timing discriminator. In backtest analysis across the confirmed call set, 5 of 7 confirmed calls showed the Compression Signal active at or before the catalyst event. It doesn't say the thesis is right. It says the spring is coiled.
The Compression Signal is displayed as a gold dot on the Scorecard.
Three independent institutional signals detected simultaneously. The rarest pattern in the framework. When it fires, someone who knows something is already positioned.
Harmonic Tension is distinct from Resonance. Resonance scores pattern strength. Harmonic Tension predicts order of events. When certain dimension combinations occur together, the framework infers which catalyst will resolve first — and which will compound afterward.
Example: A thesis where C is high but S is low suggests the catalyst may resolve without the full revaluation the sleeping giant thesis implies. The move happens, but the second leg doesn't. Harmonic Tension analysis is currently in development as part of Sextant 2.0 Phase 2.
The original Sextant scored each dimension as a static snapshot. Sextant 2.0 Phase 1, deployed March 22, 2026, adds velocity and acceleration tracking to every dimension.
Velocity is the rate of change in a dimension score between consecutive scans. A dimension moving from 4 to 6 in three weeks is more actionable than one sitting at 7 for six months. The snapshot says 7. Velocity says 6 and climbing — which means it'll be 8 next week if the pattern holds.
Acceleration is the rate of change of velocity. A dimension accelerating upward is a stronger signal than one moving at constant velocity. Acceleration identifies when a setup is entering its final compression phase before resolution. The scanner calculates both from consecutive scans per ticker.
Every score, every thesis, every published Sextant output must pass the Delbert Test. Before anything is published, it must survive being explained in plain language. If it requires jargon to be persuasive, the understanding is incomplete. Delbert doesn't care about your model. He wants to know if the company makes sense.
D (Disruption) and S (Sleeping Giant) are permanently manual. They require qualitative judgment that no automated scanner can provide. Mg-T (Management Trust) is manual for the same reason. These three dimensions are the human layer of the Sextant. They are also the dimensions most likely to catch a fatal flaw that quantitative screens miss entirely.
The Sextant dimensions do not operate independently. Movement in one dimension influences the others through the underlying market structure. Phase 3 of the framework formalizes this relationship. Documentation pending.
Custom Optimization Model — Phase 4. The framework's self-calibration layer. Rollout documentation in progress.
The Risk Modifier is a qualitative overlay applied after scoring. It does not change the GSS or Resonance score — it contextualizes it.
Thirty years working bars. Not writing about them — working them. Late nights, close calls, people at their most honest. What you learn behind a stick is that money is the clearest window into character. Watch how people behave when they're up, watch how they behave when they're down. The patterns repeat. The tells are the same whether the stakes are a $40 bar tab or a $40 million quarterly miss. First principles don't change with the setting. That's the foundation this whole thing is built on.
The Sextant Framework came from the same instinct: find the real signal in the noise. Not the consensus read, not what the sellside is saying. The question that always mattered was simpler — does this company make sense? Is there something real here the market hasn't looked at yet? The scanner automates the data layer. The judgment still has to be human. That's what Delbert would have said. That's what this is.