Trust  /  Methodology

How We Grade — Our Evidence Methodology

The grading system behind every compound page on Vallydia — explained in full, so you can judge our judgments.

---

Why this page exists

Most of the peptide and compound information online is marketing dressed as science. A vendor selling something has every incentive to call it "proven," "powerful," and "clinically validated" — regardless of what the evidence actually says.

We do the opposite. Every compound in our register gets an honest letter grade per outcome — including the grades that hurt. If a popular compound has no human data, we say so. If a mechanism is elegant but never tested, we grade the elegance and the absence separately. If the best-selling use of a compound is the least supported, our grade says exactly that.

This page explains how we arrive at those grades, so our judgments are transparent and you can hold us to them. A grade you can't interrogate is just another marketing claim.

---

The core principle: we grade evidence, not hype

Our grade answers one question: how strong is the evidence that this compound produces this specific effect in humans?

That's it. The grade is not a measure of:

• how popular a compound is,
• how exciting its mechanism sounds,
• how much a vendor wants it to work, or
• how much we might personally find it interesting.

A compound can be fascinating, widely sold, and mechanistically beautiful — and still earn an F, if the human evidence isn't there. Those are different things, and we keep them separate on purpose.

---

The two dimensions behind every grade

Borrowing the logic used by rigorous evidence reviewers, every grade weighs two independent dimensions:

1. Consistency of the evidence — do the studies agree? One promising study is a hint. Ten studies pointing the same way is a signal. Studies that contradict each other, or a single unreplicated result, pull the grade down no matter how striking the headline number.

2. Magnitude and quality of the evidence — how good is the proof, and how big is the effect? A well-powered randomized controlled trial in humans is worth far more than a mouse study or a test-tube result. A large, clinically meaningful effect matters more than a statistically-significant-but-tiny one. The strongest evidence is human, controlled, replicated, and adequately powered.

A high grade requires both: consistent findings and high-quality evidence. Strong consistency built on weak evidence (e.g. many mouse studies, zero human trials) caps the grade. A single strong trial that hasn't been replicated is likewise limited.

---

The scale: what A through F actually mean

Two things about this scale matter:

• "F" has two very different meanings, and we distinguish them in the text. One is "never tested in humans" (an unknown). The other is "tested properly and it didn't work" (a known negative). Both earn an F for "does this work in humans," but they're not the same story, and our page copy always says which.
• A grade is always tied to a specific outcome and population. "Grade A" alone is meaningless. "Grade A for the approved indication, Grade F for the popular off-label claim" is the kind of honest picture we aim for.

---

Why we grade by outcome, not by compound

This is the single most important thing to understand about our system, because it's where honest grading and marketing diverge most sharply.

A compound doesn't have one grade. It has a grade for each thing it's claimed to do.

The same molecule can be:

• Grade A for the use it's actually approved and proven for, and
• Grade F for the use it's most heavily marketed for.

Real examples from our register make this concrete:

• A hormone can have Grade A evidence for its approved medical use and Grade D/F for the "popular" social or anti-aging claims that made it famous.
• An antioxidant can be Grade A as essential biology and Grade F for the injectable cosmetic use it's sold for.
• A longevity compound can have Grade A animal-lifespan data and Grade F for proven human benefit — because mouse evidence and human evidence are not interchangeable.

If we collapsed each compound into a single grade, we'd have to pick one — and any single number would mislead. So our pages show a per-outcome table, and where a page shows a single headline grade, it reflects the compound's leading use (the primary or most-marketed application), not the highest grade in the table. We never let a compound's best-supported minor use disguise the weakness of its main selling point.

---

What pulls a grade down (deliberately)

We grade down for things marketing grades up. Specifically:

• Animal or in-vitro data only. Spectacular results in mice are common; most don't translate to humans. Preclinical-only evidence caps the human-outcome grade at the low end.
• Mechanism without outcome. A coherent, elegant mechanism is a hypothesis, not a result. We grade the mechanism's plausibility separately, and it never substitutes for clinical evidence.
• Single, unreplicated studies. One study — however striking — is a starting point, not a conclusion.
• A single strong study built on a compromised source. If the founding science has integrity problems, we say so, and it weighs on the grade.
• Marketing that outruns data. When the demand and the hype far exceed what's been tested, that gap is itself information — and our grade closes it back to what's actually shown.
• "Tested and failed." A compound that reached a pivotal trial and missed earns a lower grade than one that's simply untested — because we now have evidence of absence of effect, not just absence of evidence.

---

What we deliberately exclude

To keep the register safe and firmly on the right side of the line, our compound pages never include:

• Dosing, reconstitution, or administration instructions — for any research compound, in any form. This is a hard rule. Our pages describe what the evidence shows, never how to use a compound. (Directions for a finished, legal cosmetic product are a separate, regulated context.)
• Equivalence claims to approved drugs.
• Health or treatment claims — the register is reference information, not medical advice.

These exclusions aren't limitations of our research; they're a design choice. A reference that documents evidence honestly does not need to hand out a how-to.

---

Sources: what we weigh, and how

For every graded outcome we work from primary and authoritative sources — peer-reviewed trials, systematic reviews and meta-analyses, regulatory decisions, and the original discovery literature — in preference to secondary summaries or vendor material. Where sources conflict, we note the conflict rather than pick the flattering side. Where a claim can't be sourced, we don't make it.

The representative sources behind each grade are listed on the compound's page. We'd rather show our work than ask you to trust us.

---

Review status — an honest disclosure

Our grades are built from the published evidence, but independent expert clinical review is a separate step. Where a page has not yet been reviewed by a named, credentialed reviewer, it carries a visible "working draft — not yet independently reviewed" notice. We'd rather tell you a page is awaiting review than imply a credential it doesn't yet have. As pages are reviewed, the placeholder is replaced with the reviewer's name.

---

The bottom line

We grade evidence, not enthusiasm. We grade each use separately, so a compound's strongest point can't hide its weakest. We grade down for the things marketing grades up — animal-only data, untested mechanisms, hype that outruns proof. And we show our sources so you can check us.

If that means our grades are less exciting than a vendor's, that's the point. An honest F is more useful than a dishonest A.

---

This methodology applies to every compound page in the register. If you think we've graded something wrong — too generously or too harshly — that's exactly the kind of scrutiny this page is meant to invite.