This is the standards page for Larderlab. It states how every claim is sourced, how supplement and comparison posts earn an A-F evidence tier, how the $/g price math works, and what would change our mind. No first-person, no hype, no recipes.
Every claim cites a primary source
Each numerical claim is tied to a primary source: a peer-reviewed paper (with year, journal, and sample size), a government dataset (USDA FoodData Central for macros, FDA guidance for safety), or a published lab report (Certificate of Analysis, an independent third-party assay). 'Studies show' is banned. The bar is 'Morton et al. 2018, Br J Sports Med, meta-analysis of 49 studies, n=1,863'. Secondary summaries are never cited in place of the underlying study.
A-F evidence tiers for supplements and comparisons
Every supplement and supplement-adjacent claim gets an A/B/C/D/F tier set against the rubric in lib/content/evidence-tiers.ts. A: multiple high-quality RCTs, consistent effect, replicated in real-world settings, mechanism well-supported. B: a single high-quality RCT or multiple smaller RCTs with consistent direction; mechanism plausible. C: mechanism plausible, observational data supportive, RCT evidence mixed or limited. D: mechanism speculative, observational evidence weak, RCTs neutral or negative. F: marketed claims contradicted by published data, mechanism implausible. Creatine monohydrate sits at A on 300+ RCTs; BCAA-as-sole-protein sits at F because whole protein delivers the same amino acids at a lower $/g. Each tier ships with a written rationale and a list of PubMed source URLs.
The $/g price methodology
Buying guidance is priced in $/gram of the thing that matters: $/g of protein, $/mg of EPA+DHA, $/mg of elemental magnesium, $/mcg of B12. Every benchmark is a real price sampled from a named retailer on a dated check, typically Costco (bulk floor), Amazon (default convenience price), and the brand's own DTC store (worst-case retail). We render the as-of date, the source, and the region on every price callout. We report two decimal places or three significant figures, whichever carries more signal, and never round to whole cents.
Quarterly re-checks and the drift threshold
Prices are re-checked every quarter. When a sampled price moves more than 15% from the last benchmark, the post is flagged for a same-week update and the old figure is annotated with its date rather than silently overwritten. Tiers carry a lastReviewed date; a tier older than 12 months is re-read against current literature before it is cited again.
What would change our mind
A tier or a pick is not permanent. We will downgrade or reverse on: a published Certificate of Analysis that contradicts a product's label (mislabeled dose, undeclared filler), an independent lab failure (a third-party assay that finds the actives below spec or finds contaminants), or a new meta-analysis that moves the weight of evidence. When that happens, the affected post carries a dated revision note and the previous position stays visible in the change log.
Independence
Rankings are decided before affiliate availability is checked. We do not accept payment for placement or for a tier. We buy what we review at retail and publish the dated price we paid; we do not accept free sample packs as compensation. A brand whose evidence base earns a C gets a C even when it is an affiliate partner; a brand that earns an A is never invoiced for it. We publish F-tier coverage because 'no, this does not work' is also a question the reader came to answer.
What we will not do
Sell diet plans. Publish weight-loss numbers. Accept payment for editorial placement. Link to peptide vendors, telehealth GLP-1 prescribers, MLM 'wellness' brands, or vague 'superfood' / 'detox' products. Use 'superfood', 'detox', 'clean', or 'miracle' under our masthead.