The Method

The Method | Paul Faulkner — Forensic Intelligence Operator
The Method

We find the thesis
they built the model
to prove.

The position

Agnostic. No directional bias. No commercial relationship with any asset, sector, or operator. Long or short — wherever the asymmetry genuinely exists, that is where we go. The analysis follows the evidence. The evidence does not follow the analysis.

The job

Ask the hardest question in the room. Look where nobody else is looking. Find what the model was built to obscure. The Rogue Protocol exists in the space between what institutional research is paid to say and what the primary sources actually show. When the public record contains material the market has not priced and no private mandate exists, the analysis runs anyway.

1999
Bradford & Bingley

I designed the back-end mathematical models for the teaser rate and extreme LTV mortgage products at Bradford & Bingley. Products that would later be called liar mortgages. I understood exactly what they were because I built them — and I understood exactly what would happen when borrowers hit the standard variable rate and couldn’t refinance.

I said so. On the record. In writing. Three years before it happened.

Management proceeded. The bank collapsed in 2008. It was over.

That experience produced something that twenty-five years of institutional work has only sharpened: a profound scepticism toward any analysis that arrives pre-concluded. The institutions that failed in 2008 did not fail because they lacked data. They failed because the people with the data had every commercial incentive to reach the wrong conclusion — and the people who reached the right conclusion were not in the rooms where decisions were made.

The Rogue Protocol was built to be in a different kind of room. One without those incentives. One where the only acceptable output is what the primary sources actually show — regardless of what the narrative requires.

Institutional research doesn’t fail by accident.
It fails by design.

The most dangerous document in institutional finance is not the one that is obviously wrong. It is the one that is authoritative, well-formatted, distributed by a name you recognise — and wrong in ways that cost money before anyone notices. The Rogue Protocol exists to find those documents. Here is what they look like in practice.

Exhibit A · Public Record · 2026 JPMorgan Private Bank · Bitcoin Note · Harlap & Smith

JPMorgan published a Bitcoin note for its HNWI client base. It contained 27 forensic errors.

Not typographical errors. Structural failures: a volatility figure that was 80.8% annualised on the day of publication against the 45% cited in the note, derived from a dataset twelve months out of date. Correlation claims made without a stated timeframe. A regulatory analysis that omitted the most material regulatory development of the preceding six months. Conclusions that did not follow from the evidence presented.

The note was distributed to high-net-worth individuals making allocation decisions. It carried the JPMorgan masthead. It was wrong in ways that a first-principles read of the primary data would have caught in hours.

Consider the structural logic: if your only intelligence source on a $1.4 trillion asset class is a note from the institution managing your AUM, you are not receiving independent analysis. You are receiving content produced by an organisation with a direct commercial interest in your continued allocation — and a compliance structure that makes forensic self-criticism functionally impossible.

The forensic read: 27 documented errors. Volatility figure wrong by 35 percentage points. Published and distributed without correction. The masthead conferred credibility the methodology did not earn. Full forensic dismantling at paulfaulkner.com/jpmorgan-btc/
Exhibit B · Public Record · Feb 2026 Morgan Stanley · MARA Holdings · Initiation of Coverage

Morgan Stanley set an $8 target. The forensic read identified five structural factors they had not modelled.

The Rogue Protocol analysis, circulated to retainer clients nine days before the Morgan Stanley note was published, identified a €1.5–3.0B Exaion monopoly asset buried in transaction documents with zero sell-side coverage. Power infrastructure priced at $0.30/watt in a $1.50/watt market. 17,357 BTC encumbered — a figure no analyst had quantified, derived from footnote triangulation. Only 40% of the mining fleet profitable at $70k BTC. A 78% say-on-pay rejection mapped as a forced governance catalyst at the April AGM.

On 26 February, MARA gained +16.7% on earnings day. Not on the loss — $1.7 billion, exactly as expected. On the Starwood joint venture infrastructure re-rating. Path 2, as modelled.

The forensic read: Five structural findings absent from every sell-side report. Probability-weighted expected return: $12.35 against Morgan Stanley’s $8.00. The asymmetry was in the documents. Nobody was reading the documents.
Exhibit C · Public Record · Dec 2025 The Smarter Web Company Plc · Bitcoin Treasury · Aquis Exchange

A £200k web agency reached a £1 billion valuation on Aquis. The forensic read identified the precise collapse mechanism before it completed.

SWC was not a Bitcoin treasury. It was a regulatory arbitrage — the FCA’s ban on crypto-derivatives for retail investors had accidentally made a loss-making web design agency the only ISA-eligible Bitcoin proxy in the UK. The model depended entirely on maintaining a share price premium to NAV. The “BTC Yield” metric measured investor sentiment, not shareholder value. The TOBAM “Smarter Convert” bond’s 98% repayment clause revealed the most sophisticated capital in the room had already neutralised its risk. Once the premium inverted, the liquidity trap was mathematically inevitable.

The forensic read: Market cap collapsed from £1 billion to £127 million against £220 million in Bitcoin holdings — a 42% discount to NAV. The architecture that created the premium was the same architecture that made collapse certain once new capital dried up. It was in the documents from day one.

Full forensic dismantling (free, no paywall): Smarter Web, Dumber Money — The Bitcoin Treasury That Wasn’t →

The institutional research machine
is structurally prevented from
finding what you need to know.

Why the institution cannot see it
Commercial alignment
The analyst covers the asset their clients hold.
A bearish note on a major allocation is not a career-neutral act inside a firm with relationship exposure to that sector. The incentive to reach the comfortable conclusion is structural, not individual.
Hierarchical review
The VP cannot call out the MD’s note.
A junior analyst who spots a 35-percentage-point error in a senior partner’s published work has three options: stay silent, raise it quietly and absorb the political cost, or leave. None of these options corrects the note.
Consensus anchoring
The model is calibrated against other models.
Institutional research is validated by reference to institutional consensus. When every comparable note is wrong in the same direction, the error is invisible. The benchmark and the subject have the same flaw.
Why the independent operator can
No relationship exposure
The correct answer is always commercially neutral.
No AUM relationship to protect. No advisory mandate to preserve. No client whose allocation depends on a particular conclusion. The analysis goes where the evidence leads — long, short, or neither.
No hierarchy
The operator who reads the document is the operator who signs the work.
No junior analyst producing the first draft. No senior partner reviewing for comfort. No compliance team softening the conclusion. The person who found the error is the person who publishes the finding.
Primary sources only
The benchmark is the document, not the consensus.
Court filings. Proxy statements. Regulatory submissions. Footnote triangulation. The analysis is anchored to the primary record — not to what other analysts have said about the primary record.

Five principles.
Applied without exception.

The methodology is not a style. It is a set of hard constraints that every engagement — white paper, private mandate, retainer brief — is held to. These are not aspirations. They are the conditions under which the work is produced or not produced at all.

1
Agnostic positioning — always.

The analysis carries no directional prior. No bullish thesis to protect. No bearish narrative to sustain. Long or short, the position follows the evidence — and if the evidence supports neither, neither is the position. We are not in the business of confirming what clients want to hear. We are in the business of finding what the documents actually show.

2
Primary sources only — no proxies, no press releases.

Court filings. Proxy statements. Regulatory submissions. 10-K footnotes. The actual document, not the summary. Not the press release. Not the earnings call transcript. The document itself — because that is where the information that moves positions is buried, and it is buried there precisely because most analysts never read it.

3
Adversarial stress-testing — before delivery, not after.

Every conclusion is challenged against the strongest available counter-argument before it reaches a client. The red team is not a quality check at the end of the process. It is structural — built into every stage. The work that leaves this operation has survived the attack it will face in the market. The work that hasn’t survived doesn’t leave.

4
Speed as a structural advantage — not a compromise.

Twenty-five years of institutional pattern recognition determines which three sentences in a two-hour transcript will kill the project in six months. The AI stack — Gemini, Claude, DeepSeek, Mistral — multiplies that judgement. It does not replace it. Institutional-grade analysis delivered in days, not quarters. The intelligence that matters arrives before the market prices it in.

5
Direct access — no intermediaries, no dilution.

Large firms pitch with senior partners and deliver with graduates. That introduces noise, delay, and translation error at precisely the moment when clarity is the product you are paying for. The intelligence you receive from The Rogue Protocol is produced by the same person who reviewed your brief, read the primary documents, and signed the deliverable. No junior analyst. No account manager. No version of events between the analysis and the client.

If you want to find the real alpha, you have to look where nobody else is looking.

The sell-side covers what is covered. The consensus narrative describes what is consensus. The comfortable conclusion is comfortable because nobody with access to the information has a commercial incentive to challenge it.

The real asymmetric position — long or short — is almost always in the documents the narrative was built to distract you from. The footnote nobody modelled. The governance structure nobody mapped. The regulatory exposure nobody priced. The infrastructure asset nobody found because finding it required reading the transaction documents, not the press release.

That is where The Rogue Protocol operates. Not in the consensus. Not in the narrative. In the gap between what the model was built to prove and what the primary sources actually show. That gap is where the asymmetry lives — and it is almost always larger than the market has priced.

“The Rogue Protocol exists in the space between what institutional research is paid to say and what the primary sources actually show.”

Delivered while most operators
are still in scoping.

Traditional institutional model
Weeks in scoping before analysis begins
Junior analyst produces first draft from secondary sources
Senior partner reviews for comfort and compliance
Consensus-validated conclusions — safe, late, and priced in
Meeting notes reconstructed from memory three days later
No primary record — every version of events equally defensible
The Rogue Protocol
Senior operator reads the primary record first — AI council cross-referenced and challenged against it before any output advances
25 years of pattern recognition determines which findings matter before the stack is directed at them
Senior operator only — the person who read the documents signs the deliverable
Institutional-grade output in days — before the market prices in the finding
Primary record from the first conversation — timestamped, indexed, unchallengeable
No version of events. Only what was said, when it was said, by whom.

The question is whether
you act before the market does.

Every engagement begins with the Forensic Diagnostic. All enquiries are confidential. NDA on request. Capacity is limited.

All engagements covered by standard NDA · Client confidentiality absolute · England & Wales