YouTube trading-call claims: audit the register before trusting the result

Editorial correction: YouTube call claims need the call register

Editorial correction · 9 September 2026. The original February 13 article said that it tested 1,120 actionable calls from 42 YouTube channels and that only 8 of 42 channels beat a benchmark after costs, with a median net return per call of −0.11%. It did not link the channel universe, video and timestamp register, call-transcription decisions, standardized entry/exit rules, benchmark rows or calculation output. Those records could not be recovered from the public page or its broad references.

We withdraw the 1,120-call/42-channel cohort, channel percentages, return and predictive findings, and the categorical “most lose money” conclusion as empirical results. The original URL and publication date remain; this correction is dated separately. A channel's reach, confidence or recent streak is not treated here as evidence of skill or failure.

A proposed call-audit workflow

1. Define the universe. Record how channels are discovered, the date of inclusion, language/market scope and treatment of inactive channels.
2. Preserve the call. Store the video URL, publication time, relevant timestamp, transcript or capture, instrument, direction and any edit or deletion observation.
3. Predeclare execution. Specify the first eligible price, entry zone, invalidation, exit/time stop, holding window, position size and handling of ambiguous language before scoring outcomes.
4. Match the comparison. Choose the benchmark and return convention before evaluation; apply spread, fees, slippage and publication delay consistently.
5. Audit the result. Publish raw call outcomes, exclusions, missingness, rolling windows, uncertainty, code and an independent check. Keep the channel-level conclusion conditional on the preserved sample.

“Actionable call” should mean a predeclared set of fields sufficient to test a specific thesis. If those fields are absent, classify the item as incomplete rather than silently assigning a price or exit. The workflow is a proposed method, not a validated signal filter or trading instruction.

Source context and limits

The original references included Google's YouTube Data API documentation, Barber and Odean's All That Glitters, and Antweiler and Frank's Internet message-board study. The API documentation describes how to retrieve YouTube resources; the two papers study different populations and questions. None contains this page's 42-channel universe, 1,120 call rows, execution model or outcomes. The FCA finfluencer URL cited in the original page returned “Page not found” during the 9 September 2026 review, so it is not used to support a claim here.

Change record

Original publication: 13 February 2026. The unsupported 1,120-call, 42-channel, return and predictive findings were withdrawn and the categorical title reframed as neutral call-audit guidance: 9 September 2026. The original wording remains in version history and is not promoted as current research. AI-assisted editorial review; no independent human expert review.