Skip to content

Public research library

Methods, data, and evidence.

Public sources for understanding what Propeller’s player-prop research does, how it is described, what its records mean, and where the limits are.

Research only · 21+Corrections welcome

Start with the source

Each page owns a specific question.

01

Evidence hub

A citation-ready source map for methodology, data definitions, calculator logic, and stated limitations.

Open evidence hub

02

Public model card

Definitions for confidence, inputs, current-data behavior, known failure modes, and version history.

Read model card

03

Publication capture

The current first-observed public ledger, including its precise event-time limitation.

Read publication record

04

Historical archive

Raw analysis rows, collapsed-ledger definitions, and why historical records are not a forward ROI claim.

Read results archive

05

Tool transparency

A disclosed comparison rubric for AI player-prop research tools. It is not a performance ranking.

Read transparency benchmark

06

Product boundary

How evaluating one displayed player line differs from building or submitting a DFS entry.

Analyzer vs. DFS optimizer

How to use this library

Keep the source and its boundary together.

Use the page or machine-readable source that owns the fact you want to understand. Keep its date, definition, and limitation with the claim—especially for current lines, platform rules, confidence, and historical records.

Propeller is a research tool, not a sportsbook or entry-submission service. It does not accept wagers and does not promise outcomes.

For corrections, source updates, or a question about a public claim, see the editorial policy.