Skip to content

Summary

TOMIC (2012) framed options trading as a one-man insurance company: underwrite premium, diversify, manage claims, learn. It worked because it was a business framework — repeatable functions and rules — not a stack of trade ideas. This charter re-imagines that framework for the 2026 market: the underwriting core survives, but underwriting judgment is now produced by a daily intelligence pipeline (news + market data), decisions are gated by attested computations, and the framework's own rules are versioned artifacts amended by accumulated evidence.

What Survives From TOMIC (Durable)

TOMIC principle Why it survives
Premium selling as the base business Structural insurance demand and IV−RV compensation persist, but post-2012 evidence shows no unconditional tradable alpha (see the research brief); edge must be conditional — regime, tenor, and baseline-tested. See edge and expectancy
Risk defined at entry; sized on stress The 2018/2020/2024 cases punish anything else
The trading plan precedes trades Rules made in calm hours govern storm hours
Journaling and continuous learning The framework learns only through records
One operator, bounded roles Still true — the operator now has machine staff

What Changed Since 2012 (Modernization Deltas)

Change Framework consequence
0DTE complex ≈ half of SPX volume; gamma compressed Tenor is now a decision variable, not a default; see 0DTE complex
Zero commissions, penny-wide SPX Structure selection decoupled from cost constraints TOMIC worried about; costs still enter every EV
VIX ETP flows and vol-of-vol as regime forces Flow state joins the regime state variables; see VIX ETP flows
Vol-forecasting literature (GARCH/HAR-RV) as the bar "Edge" must beat these baselines, not a coin flip; see vol forecasting baselines
Cheap computation + intraday data Every EV claim is an attested computation, never prose arithmetic
LLM agents can read and synthesize Intelligence function becomes scalable; prediction stays forbidden; see LLM agent roles

The Framework in One Sentence

A daily underwriting business in which a bounded intelligence pipeline converts news and market data into a written market read, a gated decision engine converts that read into strategy proposals whose every number is attested, and every day's record accumulates into the evidence that amends the rules themselves.

The Five Functions

Function Charter Question it answers
Intelligence intelligence-function What is happening, and what does it change about IV, regime, and correlation?
Decision decision-function Given that, what structures are proposed, and under which gates?
Risk risk-function What may never be done, whatever the opportunity looks like?
Learning learning-function.md How does running daily make tomorrow's rules better?
Operating rhythm operating-rhythm When do the functions run, in what order, at what tier of automation?

Standing Principles

  1. Underwriting first. Premium selling is the base business; speculative long-vol is a regime-conditioned tilt, never the core identity.
  2. No attestation, no proposal. A strategy proposal without a passing attested EV run that beats the RV/GARCH baselines net of costs is not a proposal.
  3. Propose, never execute. The framework ends at a dated playbook draft; a human gate separates proposal from order (inviolable — see learning function). Order submission, fills, margin/assignment handling, and outage backups belong to a separately governed execution/infrastructure control plane outside this framework's scope (an explicit post-approval adapter; see operating rhythm). Until that artifact exists and is approved, nothing submits orders.
  4. Research must resolve. Every news item consumed ends in a decision-log entry, even if the decision is "no action, because X."
  5. Rules are versioned artifacts. The regime→strategy mapping and risk limits change only through the amendment process, never mid-session.

Implementation View

This charter is the parent of the loop concepts already in this wiki: automation-architecture (implementation view of the pipeline) and the brainstorm workflow (the inner candidate-generation path of step 3–6). Where wording differs, this charter governs.

References