How we check: calculator tests and seller rules

1. How the calculator works

It takes three numbers you type (peptide in the vial, water you added, amount you want in one draw) and does four bits of arithmetic:

If you type mcg, we divide by 1,000 to get mg first. The "100 units = 1 mL" rule is the U-100 insulin syringe standard (ISO 8537). A 1 mL syringe reads to 100 units, a 0.5 mL to 50, a 0.3 mL to 30. If your draw is past the end of the syringe you picked, the calculator says so.

The size buttons only fill in the vial amount. There is no suggested amount for the draw and no suggested amount of water, anywhere.

The three tests

Before any update goes live, the calculator code is run against these three cases. If one fails, nothing is published.

Numbers are shown rounded (four decimal places for mL and strength, two for units and draws) but the math underneath is not rounded between steps. Nothing you type is sent anywhere.

2. What we leave out, and why

Some things are missing from this site on purpose. Saying so is better than pretending.

• Amounts of our own. PeptideGauge never proposes an amount, a schedule, or a starting point, and the calculator only converts the number you type. One narrow exception in form, not in substance: clinician commentary sections quote amounts that a named clinician states on camera, labeled with their framing and linked to the video. Those are reported words, not our guidance. How that layer works.
• How much water to add, or which water. No brand names, no volumes, no step-by-step mixing instructions. The water amount in the calculator is something you type in, never something we suggest.
• Medical advice about prescription drugs. Some pages now cover compounds that are approved medicines or are in large trials. We report the published research and what clinicians say about them in public. We do not tell anyone to take them or how; that belongs with a doctor who knows you.
• Telling you where to buy. We are not a store and link to no cart. When the seller list is published it will rank on paperwork, not on price or on anyone's word.

3. How we rank peptides on a goal page

Each goal page puts peptides in an order. The order is arithmetic, published here before you see any of the results, and every number that feeds it is printed in the table itself so you can redo the sum or decide the weights are wrong.

Four scores out of 100, added with fixed weights:

• Evidence strength, 40 percent. The strongest single study we found for that goal. Randomised trial in people 100, systematic review 75, small human study 50, review article 30, animal study 25, cells in a dish 10.
• Human data, 25 percent. How many of the studies for that goal were run in people. None 0, one 40, two 60, three 80, four or more 100.
• Safety record, 20 percent. An FDA approved drug with a full label and years of market surveillance scores 100. Some human safety reporting 70, animal safety data only 35, no human safety data found 10.
• Regulatory clarity, 15 percent. FDA approved 100. In FDA trials but not yet approved, 60. On the FDA list of compounding nominations that were withdrawn 25. If the FDA also states it found no human exposure data, 10.

A peptide is only ranked once we have read its studies and linked each one. That is why the lists are short. If we change a weight or a scale, the pages are rebuilt from the new rules rather than edited by hand.

Every study behind every ranking is in one place: the evidence table.

Scale change 08-26-2026: added the approved drug safety category and the in trials regulatory category when an approved drug and an investigational drug joined the goal pages. No previously published score changed.

4. How we judge sellers

The seller table is live. These rules were set before it existed, so the scores cannot be bent to fit a result. The scoring formula is printed under the table itself.

Score and row order are two different things. Every score in that table is what the formula produces and nothing else. The order the rows appear in is an editorial choice: we place our own pick first, then everything else by score. The table says so above the rows, and the picked seller still shows its real score, even when sellers below it score higher. If you only trust the formula, read the score column and ignore the order.

• Only what the seller's own pages show. Do they post a lab report (a "COA") for each product? Do they name the outside lab that ran it? Can the batch number on the report be matched to a product batch? Does the report say what was tested and how? Is the report dated?
• Every answer carries a date and a place. Each cell says the day we looked and the page we looked at. If we could not check something, the cell says "not verified." It never guesses.
• No piling on. We do not say anything negative about a seller beyond what their own pages show or fail to show. A missing document is recorded as missing, not as an accusation.
• Re-checked on a schedule. The list shows when the last full check was done.

5. Where our facts come from

Pages about what a peptide does (coming later) will cite published research by its PubMed or PMC ID for every claim, and say what kind of study it was (cells, animals, or people; how many; what year). Seller pages, forum posts, and second-hand summaries do not count as evidence. A claim we cannot source gets cut, not softened.

Pages are signed by their editor, Mark Sommers, who is named on the About page. He is not a doctor and holds no medical licence, so nothing here has been reviewed by a clinician. That is the reason this site reports what a study found and never reports an amount.

4b. How we grade a study

On each peptide page every study gets one tag, based on who or what was studied, not on what the authors concluded:

• Cells in a dish. Lab work on cells or tissue. Shows a mechanism might exist. Says nothing about people.
• Animals. Mice, rats, pigs. Useful for hints; most things that work in rodents do not work the same way in people.
• Small human study. A case series, pilot, or open-label study with no comparison group, usually under 30 people. Cannot separate the treatment from chance, placebo, or time.
• Human trial. People randomly split into treated and untreated groups. The only tag that can show an effect is real. We still note the size; a 9-person trial is a start, not proof.
• Review and Systematic review. Papers that sum up other papers. A systematic review follows a stated search method; a plain review is one author's reading. We flag it when a review's author discovered or sells the peptide.

Each component also gets a one-word human evidence line: None (no human studies found), Limited (only small studies or a single trial), Moderate (more than one randomized trial in the form that is sold). The count next to it is how many human studies our PubMed search turned up on the date shown. Every line links to its PubMed record so you can read the abstract yourself.

6. The clinician commentary layer

Added 08-26-2026, widened later the same day. Some peptide pages carry a section called "Clinician commentary (YouTube)". It reports what four clinicians with large audiences say on camera about a peptide or a blend: Dr. Alex Tatem, Dr. Ashley Froese, Dr. Jones DC, and Dr. Craig Koniver. Each point links to the specific video it came from, with the title shown, so you can watch the source instead of taking our word for it.

Three of the four publish on their own channels. Dr. Koniver does not; he appears as a guest on other people's shows, so his videos are credited to the show that recorded and published each interview, and we name that show next to the title.

• Commentary is not evidence. Nothing in that section changes a study grade. The graded research on each page stays the record of what is known, and the two never mix.
• Amounts are quotes, not guidance. Where a clinician states an amount or a schedule on camera, we report it with a label saying how they framed it: trial dosing, community practice, a wellness clinic habit, or a personal anecdote. PeptideGauge does not recommend amounts. See what we leave out.
• Rankings stay personal. Dr. Tatem publishes a letter tier list on his channel. Where a tier appears here it is labeled as one clinician's ranking. It is not a consensus and it is not our rating.
• Quotes come from captions. Some blocks carry a short quoted line. Those are taken from the automatic captions YouTube generates, not from a hand transcription, so the wording can differ slightly from the audio. We say so under every block that has one, and the link lets you hear the original.
• We do not print everything. The first two videos per clinician are open on the page. Any others fold into a "more videos" toggle in the same section. Nothing is hidden from search engines or from readers without JavaScript; the toggle only stops commentary the site itself grades as secondary from crowding out the research above it.
• Who said it matters. These are interviews and monologues, and the host is not the guest. Where a claim came from an interviewer rather than the clinician, it is not in this section at all.
• How it was built. We keep notes and transcripts of the cited videos on file, off the website. Every video link, title and channel was checked against YouTube on 08-26-2026; all forty resolved and every one was published by the channel we credit. Every quoted line was checked character by character against the transcript of the video it is attributed to. The paraphrase is ours; the claims belong to the named clinician, and the link lets you check the paraphrase.