Our scouting prose is machine-composed. Here are the rules.

Three labels, three different promises.

• Prospect brief: facts and model context only; individually attributed observations may appear, but are not called consensus.
• Sourced scouting consensus: at least three independent analyst voices including an anchor/high-authority source, or two independent anchor/high voices.
• Draftanomics film review: original observations across at least three referenced games. This label cannot be earned by scraped prose or statistics.

Multiple articles from one analyst remain one voice. A player with identity facts and nothing else still gets a clearly marked limited brief, never a fabricated qualitative evaluation.

A master report is a depth designation, not a film-review label. It means a dated, player-specific research pass reconciled primary records, multiple independent evaluations, dissent, availability and public football context into one versioned house report. It still remains a sourced scouting consensus unless Draftanomics logs its own game-referenced film work.

What the composition pipeline does.

A Draftanomics report is one house read of every scouting report we have actually found on the player, checked against the statistical record. A player-first engine hunts public writeups across the sources we trust, keeps only usable prose, and scores the file: rich, adequate, thin, or empty. Confidence follows the file, not the board slot — player 5 with four independent reports is a rich file; player 276 with one blurb is a thin file, and the prose says so. Two or more independent reports become a master synthesis. One report is a single-source brief. Zero reports is a numbers-only brief with no film claims. The original quotes stay on the page, attributed, under “In their words.” Before a word is drafted, the pipeline builds a deterministic fact sheet — the prospect's consensus standing, production, athletic testing, age and program context, the full reports on file, plus every scouting observation we've ingested, each tagged to a named scout at a named outlet (the full roster is on /sources). Validators then compare the draft with the same facts. We do not invent film we have not watched.

The hard rules.

Every draft passes through a validator before publication, and the validator is not a spell-checker. It checks four publication rules:

• Film language requires a source. Any claim about how a player looks on tape — bend, hand usage, route feel, play strength — must resolve to a retained observation from a named evaluator. No source, no sentence. Automated reports may show sentence-level footnotes; master reports publish clean house prose with a linked source manifest that preserves the research trail.
• Every number must be in the data. A forty time, a rank, a snap count, a dollar figure — each one is checked against the live fact sheet. A number the data can't produce is a rejected draft.
• No hedge-stacking. “Could potentially perhaps develop into” is how prose hides that it has nothing to say. The validator rejects it.
• No template constructions. Reports that read like a mail merge — the same skeleton with the name swapped — are rejected as a class.

Rejected qualitative drafts never publish. The page falls back to the deterministic prospect brief, whose sentences are assembled directly from verified fields and cannot introduce film claims.

What we don't do.

Unless a report is explicitly labeled Draftanomics film review, we do not imply that Draftanomics performed an original film study. A film-derived conclusion in a sourced or master report is our synthesis of the evaluators linked in its research trail. We do not invent observations to round out a profile, turn rankings into traits, or convert an identity-only file into a qualitative evaluation. Thin files say exactly what is verified and exactly what remains missing.

Why we work this way.

Draftanomics's whole pitch is receipts. The site's methodology is public — including which prior claims were withdrawn, a record kept at /methodology — and the writing is held to the same standard. Machine composition is what makes that standard enforceable — every claim in every report is traceable to a source or a stat, checked by a process that doesn't get tired and doesn't get attached to a take. We'd rather tell you exactly how the sausage is made than ask you to trust a byline.

Who we cite and what data we build on: /sources. Questions or corrections: hello@draftanomics.com.