How we review, code, and rate.

1. What we review

We review trading strategies that have been publicly taught by a named educator, trader, or influencer — meaning the rules have been disclosed in free livestreams, free YouTube content, public X/Twitter posts, podcasts, books, or paid courses where the core ruleset has been described on-record. If a strategy has never been publicly explained, it is not eligible for review.

2. How we source the rules

We work exclusively from public material. We do not pirate paid courses. We do not accept leaked content. Our coverage relies on what the educator has already said in the open — clips, slides shown on stream, written posts, podcast transcripts, or our own paid enrollment in a publicly-available course where the educator has consented to the strategy being discussed.

For each review we maintain an internal source log — every rule we coded is mapped to a specific public citation (timestamped clip, post URL, slide screenshot). The log is not published, but it is the basis for our right-of-reply process below.

3. How we reimplement

We write our own code. We do not redistribute, decompile, or copy the educator's code. Where the public material specifies parameters (e.g., "I use a 20-period EMA on the 1-minute chart"), we use those exact parameters. Where the educator describes a concept but not the parameter (e.g., "I use a short EMA"), we note the gap and select a value consistent with the rest of the rules; the rating page discloses which parameters were specified versus inferred.

Where a strategy has multiple variants the educator has discussed across different sessions, we code the most recently described version and note prior variants in the commentary.

4. How we backtest

Backtests run on QuantConnect against minute-level historical data with realistic commission and slippage assumptions modeled per instrument. Standard test window is five years where data is available, ending at the parameter-match date stated on the review. We do not curve-fit; the parameters in the backtest are the parameters the educator publicly taught, not parameters we optimized in-sample.

Each backtest publishes: equity curve, Sharpe ratio, max drawdown, win rate, expectancy, average trade duration, and parameter-match attestation. Hypothetical performance disclosure per CFTC Rule 4.41(b) applies in full; see the disclosure page.

5. The rating

Every reviewed strategy receives a single rating on a 1.0 to 10.0 scale, scored to one decimal place (e.g., 6.1, 3.5, 8.7). The decimal resolution is deliberate — a binary or three-tier scale cannot honestly distinguish a strategy that backtests slightly better than another. The rating is our editorial opinion, based on the backtest result, the robustness of the rules under parameter variation, the consistency of the educator's public description with what actually trades, and the realism of the assumptions required to make the strategy work.

The rating is opinion; the backtest data is fact. Both are published side by side so readers can disagree with our judgment without disputing the numbers.

6. What we publish — and what stays gated

Free, public review: the educator's name, a concept-level summary of the strategy, the backtest headline metrics, the rating, and our commentary. Enough for a reader to understand whether the strategy is credible.

Gated: the exact parameter set we coded, the precise entry/exit triggers as implemented, and live signals from strategies we rate highly. These sit behind the paid Signals tier (live calls) and the premium tier (full mechanical reveal). We gate the mechanics for two reasons: it is what sustains the desk financially, and it prevents bad-faith actors from cherry-picking rules out of context to mislead retail traders.

7. Right of reply

Before publishing any critical review, we contact the educator at least 72 hours in advance with the headline finding, the parameter-match summary, and an invitation to respond. We publish their response — in full, lightly edited only for length and readability — alongside the review. If they decline or do not respond, we note that plainly. If they identify a factual error in our reimplementation, we hold publication, correct, and re-test.

8. No pay-to-play

We do not accept payment, sponsorship, affiliate revenue, or any other compensation from an educator in exchange for: covering them, omitting them, altering a review, delaying publication, removing a published review, or upgrading a rating. This is a hard line. Coverage decisions are made editorially.

Where a relationship exists that a reasonable reader would want to know about — for example, the operator has a prior personal relationship with the educator, or has licensed a strategy from them for a GoyBot paid product — the review carries a clearly labeled conflict disclosure at the top. See also our affiliate disclosure.

9. Reviews and products are separate

Our editorial reviews and our paid algorithmic products are operated as two independent surfaces. Reviewed educators are not GoyBot products. GoyBot products are not named after reviewed educators. Where a GoyBot paid algorithm is licensed from a named educator, that relationship is disclosed on both the product page and on any review of that educator's other work.

10. For educators: right of reply, takedowns, partnerships

If you are an educator whose strategy we have reviewed and you wish to respond, correct an error, or discuss a licensing partnership, email legal@goybot.com. We respond to all such requests within five business days.

11. Changes to this policy

We may revise this methodology as the desk evolves. Material changes will be noted on this page with a dated changelog entry. The current version dates from the timestamp at the bottom of this page.

Version 1.0 — published 2026-05-31.