How the Evidence score works
The Evidence score is a 0-100 read on how much real evidence stands behind a compound, and where it sits with regulators. It is built from four things, each capped at a fixed number of points, and the same way for every compound. No part of it is a guess, and no vendor can pay to move it.
A high score means the evidence is strong and does not mean anyone should take it. A low score means the evidence is thin and does not mean a product is unsafe. It is not a safety score, not a purity or quality rating, and not medical advice. It is not a claim that any compound treats, cures, or prevents any condition, and not an endorsement or a recommendation to buy or use anything. It measures one thing only: how much legitimate evidence exists.
The four things we measure
Each compound earns points in four categories. Add them up and you get the score out of 100.
Regulatory standing
0 to 25 ptsWhere the compound sits with the FDA.
| FDA-approved prescription medicine | 25 |
| In clinical trials, not yet approved | 15 |
| Status unclear | 7 |
| Sold as a research chemical, not approved | 5 |
Human-evidence grade
0 to 35 ptsThe strongest tier of real evidence we can find, hand-graded. Human data counts for far more than animal or anecdote.
| Controlled human trials | 35 |
| Some human data, not conclusive | 22 |
| Animal or lab-dish only | 9 |
| Community reports, no controlled data | 4 |
Registered human trials
0 to 25 ptsPulled live from ClinicalTrials.gov, so it is objective and updates on its own. More trials and more advanced trials score higher. Zero registered trials scores zero.
| 50 or more registered studies | up to 18, before phase bonus |
| 10 to 49 studies | 14, before phase bonus |
| 3 to 9 studies | 9, before phase bonus |
| 1 to 2 studies | 5, before phase bonus |
| Most advanced phase (added on top) | +1 to +7 |
Cited sources
0 to 15 ptsHow many references back the profile, and whether any are primary or regulatory (PubMed, PMC, journals, .gov, the FDA).
| Five or more sources | 15 |
| Three to four sources | 11 to 13 |
| One to two sources | 5 to 8 |
| A primary or regulatory source is present | +2 |
| No sources | 0 |
What the number means
Strong, human-grade evidence
A real evidence base, with gaps
Early evidence, mostly unproven in humans
Very thin human evidence
What we leave out, on purpose
How many papers exist
A compound can have hundreds of animal papers and zero human proof. Rewarding sheer volume would reward the exact hype this score exists to cut past, so paper count does not move the number.
FDA recalls
A recall is a manufacturing or safety event on one batch. That is a different question from how much evidence backs the compound, so it is shown on the profile but kept out of the score.
Purity
Whether a specific vendor's batch tests clean is the job of Verify, our separate purity index. The Evidence score never claims to know what is in any particular vial.
Sources we cite
Every profile is built on public, checkable sources, never vendor marketing copy. The live data layers refresh on their own so profiles do not go stale.
ClinicalTrials.gov
The US registry of human trials. Drives the registered-trials part of the score and the trials shown on each profile.
openFDA
The FDA's open data, used to surface any recall or enforcement record tied to a compound.
Europe PMC
A large index of life-science literature, used to show the weight of published study behind a compound.
PubMed, PMC, journals and the FDA
The primary and regulatory references cited on each profile, the same ones a clinician would check.
Why we take no vendor money
The whole point of PLOBT is a read you can trust because nobody paid for it. No vendor can sponsor a compound, buy a higher score, or pay to be ranked. The score is computed the same way for everything, from an FDA-approved medicine to a research chemical with no human data. If that ever changes, this page will say so plainly.
Everything here is educational information, not medical advice, and not a recommendation to use any substance. Many compounds described are not approved by the FDA. Always talk to a qualified healthcare professional before making any health decision.