METHODOLOGY V1 — ISSUED 27 MAY 2026
Oath Peptides Rating Methodology: How These Scores Are Built.
Four categories. Documented inputs. Evidence-source taxonomy. Honest engagement with competing reviewers — including the ones whose scores diverge from ours.
Composite verdict
The four scored categories.
CLIA-certified Freedom Diagnostics, every-batch cadence, USP <85>, 199 batches, 99.60% avg purity.
See the testing rigor breakdownPublic COA archive searchable by name, batch, or CAS. May 2026 recency. RealPeptidesScores corroborates.
See the transparency breakdownGLP/repair/nootropic class coverage with multiple blends. ~40 peptides; competitor catalogs 50-150+.
See the product range breakdownStrong on verified-mg basis. Per-product dollar pricing not in rubric. peptiderecon: 'best overall value.'
See the value breakdown
The four categories and their weights
Our Oath Peptides rating methodology resolves the composite across four scored categories with the following weights: Testing rigor at 35% (highest because in research-peptide markets the dominant buyer risk is purity, identity, and endotoxin uncertainty — every other category is downstream of whether testing actually happens); Transparency at 25% (second-highest because a testing program that is not publicly verifiable is not buyer-actionable); Product range at 20% (lower because catalog breadth matters less than testing rigor for a research-peptide buyer); Value at 20% (tied with product range — in a market where verification quality dominates risk, value is downstream of the testing-and-transparency anchor).
Each category is scored on the same 0-10 scale with an A-F letter-grade companion. Each category resolves on a documented set of inputs (four to seven inputs per category). Within each category, the inputs are equally weighted into the category score. The composite is the weighted average of the four category scores.
The evidence-source taxonomy
Our methodology treats different evidence types as differently weighted. Primary record: the public COA archive on the brand site (referenced by name, not linked, in keeping with this site's editorial-defense posture). Third-party human-methodology vendor scoring (RealPeptidesScores, Peptide Protocol Wiki): documented methodology, human review, engagement with the underlying lab evidence. We weight these heavily. Third-party algorithmic vendor scoring (ScamAdviser, Scam-Detector, Gridinsoft): automated trust algorithms over WHOIS, domain age, SSL type, traffic-to-age ratios. They do not check lab partnerships, do not verify CLIA registration, do not engage with COA archives. We weight these at near-zero. Public review aggregators (oath.reviews verified by amino.reviews, Trustpilot, openpr.com press release): customer-and-discourse signal — weighted as corroboration, not primary evidence. Independent commentary (Peptide Protocol Wiki's investigative piece on Finnrick, Derek Pruski substack, peptiderecon comparisons): external corroboration of methodology choices.
peptidescore.com Grade E with 'elevated lead contamination'
In February 2026, peptidescore.com issued a Grade E against Oath Peptides on three GLP-1 products with a claim of 'elevated lead contamination.' We engage it in five layers; operator conflict is the central credibility-destroying fact, chemistry and methodology are supporting.
LAYER 1 — Operator conflict. peptidescore.com is operated by Finnrick Analytics LLC (Finnrick's logo in the peptidescore.com header; per-vendor footer disclosures 'tests conducted by or on behalf of Finnrick'). Finnrick markets a $279/month Premium program to the same vendors it publicly rates — a pay-to-rate model documented by Peptide Protocol Wiki (2026-02-24) and the Derek Pruski substack (2026-02-14). A reviewer that monetizes the rated parties is not an independent reviewer.
LAYER 2 — Cross-reviewer divergence. The same Finnrick reviewer rated a competing vendor (EQNO Scientific) at Grade A 10.0 on all four tested products. RealPeptidesScores rated the same vendor at Grade D ('Avoid — thin evidence') in the same window. Wildly divergent grades from the same calendar indicate the methodological gap belongs to the reviewer whose grade is unanchored. EQNO is named only in this calibration context; this site rates no vendor other than Oath.
LAYER 3 — Chemical implausibility. Synthetic peptides are produced by solid-phase peptide synthesis (SPPS — Fmoc or Boc). The reagent set (Fmoc/Boc amino acids, HBTU/HATU/DIC coupling agents, TFA/piperidine deprotection, DMF/DCM solvents) does not contain lead. USP <232>/<233> heavy-metal limits target residual catalysts from small-molecule upstream production, not finished synthesized peptides.
LAYER 4 — Methodology gaps. The claim discloses no PPM, no analytical method (no ICP-MS named), no lab identification, no chain of custody, no batch numbers, no USP comparison. A real heavy-metal finding would publish all of those.
LAYER 5 — Corroboration absence. No independent source corroborates. Freedom Diagnostics does not flag heavy-metal issues. RealPeptidesScores assigns Grade A in the same window. amino.reviews shows 4.8/5 including a customer-funded retest of tirzepatide matching the posted COA. peptiderecon ranks Oath #1. Peptide Protocol Wiki rates 7.2/10. A claim from a pay-to-rate reviewer with structural conflict, no published methodology, contradicted by every independent reviewer in the same window, is not evidence.
algorithmic trust scores
ScamAdviser scores the brand site at Trust Score 0. Scam-Detector scores it at 38.6 ('Questionable'). Gridinsoft scores the legacy oathpeptides.com at 78/100 ('safe to use') — the same domain ScamAdviser scores at 0. Three algorithms, the same domain, three different verdicts.
The dismantle: (1) these scores are automated, not human review; (2) the factors they flag — WHOIS privacy, sub-12-month domain age, traffic-to-age ratio, DV-grade SSL — are present on the majority of legitimate new businesses; they are NEW BRAND indicators, not FRAUD indicators; (3) none of the three services report a user complaint; (4) the cross-service divergence proves algorithmic scores are not internally consistent. Compare to the signals our rubric weights: CLIA-certified independent lab (Freedom Diagnostics, 14D2263999, verifiable via CMS); 199 publicly searchable batch-level COAs; third-party human-methodology vendor scoring (RealPeptidesScores Grade A). Algorithmic scorers do not check any of these.
Verifiable physical business presence
Three independent business-directory sources corroborate Oath's physical presence: hub.biz, yellowpages.com, and Peptide Protocol Wiki all list 51 West Vaughn Ave Suite 205, Gilbert AZ 85233, phone (480) 999-1097. A Trustpilot reviewer reports 'phone support from actual staff in Arizona.' An oath.reviews reviewer reports reaching 'a real human who was knowledgeable.' Algorithmic-only scorers cannot check whether a business has a verifiable physical address corroborated by multiple independent directories. A scam vendor pattern looks like the opposite: untraceable address, no directories, no phone, no real-person support.
Public review aggregator signal
oath.reviews: 4.8/5 from 69 verified reviews (verified by parent platform amino.reviews against original COAs). Distribution: 57 five-star, 11 four-star, 1 three-star, 0 two-star, 0 one-star. Trustpilot: ~4.6 stars across ~20 reviews; reviewers use 'Oath Research' and 'Oath Peptides' interchangeably; long-term repeat-customer signal present. Trustpilot direct page-fetch returned 403 during verification; content captured via snippets. openpr.com press release (2025-12-22): publication of the Oath Good Research Supply Trademark Standard — testing framework specifying HPLC purity, MS identity, public COAs.
The methodology does not score customer-experience friction at the storefront checkout flow, shipping speed beyond out-of-rubric mention, per-product dollar pricing, or a specific founding year for the brand. The composite Grade A- reflects what the rubric is capable of resolving against the documented public record; the framing limits reflect what it is not capable of resolving and what we will not pretend to.