Source Index
29 · Source Index
Section family: provenance & methodology · The bibliography behind every claim in this guide.
This is the front door to the project's source layer. The machine-readable catalog of every source we drew on lives in two files described below — this README explains what's in them, how the sources were vetted, and how a citation anywhere in the knowledge base traces back to one of them.
Everything else in this guide tells you what the premium-selling approach does and why. This section answers a narrower, load-bearing question: where did each statement come from, and can you check it yourself? A trading guide that cites nothing is just opinion. So every substantive or quantitative claim in the knowledge base is tagged inline with an evidence label and, wherever possible, a real URL — and all of those URLs are registered, de-duplicated, and graded here.
---
The two files of record
The full registry is intentionally kept as data, not prose, so it can be sorted, filtered, validated, and regenerated:
- ./SOURCE_REGISTRY.md — the human-readable registry. 105 catalogued sources, of which 74 are web-verified (a cataloging agent fetched the URL and confirmed it resolves to the expected content). Each row records title, type, difficulty, verification status, date, author, topic, and key concepts, grouped by category.
- ../_data/sources.json — the same 105 sources as structured JSON (`count: 105`, `verified: 74`), one object per source with a stable `id`, `url`, `url_verified` flag, `type`, `topic`, `concepts[]`, `difficulty`, `author`, `date`, and `category`. This is the file the build tooling reads; the Markdown registry is generated to mirror it.
If the two ever disagree, `sources.json` is authoritative — the registry is its rendering. Both were compiled through live web research and are maintained as the project's single bibliography.
---
The six source categories
Every source is filed under exactly one `category`. The split reflects what kind of evidence the source is, which matters when you weigh a claim:
The category is descriptive, not a quality score — a `learn-center` page can outrank an `original-shows` clip for a given fact, and vice versa. Quality is judged per claim by the evidence taxonomy below.
---
The evidence / verification approach
Two distinct things get checked, and it's worth keeping them separate:
1. Does the source exist? (verification) The 74 web-verified sources were fetched live and confirmed to resolve to the expected content. The remaining 31 are recorded by name with a best-known URL and flagged for QA re-verification — they are not deleted, because a known-good title with a stale link is still a real source; it just hasn't been re-confirmed. Verification status travels with each record (`url_verified` in JSON, the Verified column in the registry). Spot-checks for this index confirmed, for example, that the IV Rank & Percentile concept page states selling premium is relatively more attractive above an IV Rank of 50 , that the Managing Winners page documents the ~50%-of-max-profit exit , and that the Options Greeks reference covers delta, gamma, theta, and vega together .
2. How much does the claim deserve? (grading) Verification says the link works; it says nothing about whether the underlying assertion is backed by data. That is the job of the evidence taxonomy — the same three-axis label used everywhere in the guide:
- Grade is provenance: A = a published, dated options-education study with a real backtest; B = house canon stated and repeated across the source material; C = standard option theory true independent of any single methodology; D = opinion or unverified third-party. A third-party source is at most Grade C — only a genuine first-party study earns Grade A.
- Confidence is how sure we are of the magnitude or applicability (High/Med/Low).
- Research-backed vs Heuristic flags whether a number comes from a study or from sensible rule-of-thumb.
The full grading rubric, with worked examples, lives in 18_research_findings — that section is the per-claim citation layer; this section (29) is the per-source registry it points at. Read them as a pair: 18 holds the findings, 29 holds the sources.
---
How citations across the knowledge base map back here
Every section README and strategy page carries a `sources[]` list in its front-matter, and every inline evidence tag ends with a `src:` URL. Those URLs are not free-floating — they resolve to entries in this registry:
1. A claim in, say, 05_trade_management or 09_strangles is tagged inline with its grade and a `src:` URL.
2. That URL matches a record in ../_data/sources.json (same `url`, with its `url_verified` flag and `category`).
3. The record renders as a row in ./SOURCE_REGISTRY.md, where you can see the source's type, author, date, and concept coverage at a glance.
This lets a reader audit any claim end-to-end — from the sentence, to the source, to the verification status — and lets the build tooling check that no citation points at a URL the registry doesn't know about. Sources here underpin the whole guide, from 00_foundations and 03_implied_volatility through the strategy sections — 07_short_premium, 10_iron_condors, 11_credit_spreads — and the mechanics and platform sections 22_mechanics and 23_platform_usage. The cheat sheets in 27_master_cheat_sheets and the definitions in 28_glossary draw on the same pool.
A standing principle: prefer primary sources. Where the study itself is published, cite the study; third-party summaries (a handful of explainer sites such as `daystoexpiry.com`, plus CME and podcast entries) are kept for context and convenience but never outrank the primary source, and never qualify as Grade A.
---
_Evidence-labeled per the Project Charter. Education only, not financial advice._