It's all just placebo because the studies are so small

The objection conflates two distinct concerns and applies them to a corpus they do not uniformly fit. Concern one: small studies have wide confidence intervals around effect estimates, are more vulnerable to publication bias, and are more easily contaminated by allegiance effects from the investigators who run them. This concern is real and applies to the small-cohort end of this corpus — the rabbit-tendon BPC-157 papers, the 60-patient Russian Selank trial, the small mechanistic Semax studies. For those peptides, the effect-size estimates carry meaningful uncertainty even where the trials themselves are competently executed. Concern two: placebo response in healthy or near-healthy subjects can be substantial, particularly for outcomes that are subjective (anxiety, sleep quality, perceived recovery), expectancy-driven (cognitive performance, sexual response), or operate on time-courses where regression to the mean does most of the work. The objection — "it's just placebo" — is gesturing at this concern. The two concerns are not the same thing, and conflating them obscures the right response. For the peptides on this list with regulatory-grade evidence — Tesamorelin (n=412 NEJM Phase III), Semaglutide (n=1,961 NEJM Phase III plus the rest of the STEP program), Tirzepatide (n=2,539 SURMOUNT-1), Retatrutide (n=338 NEJM Phase 2 plus ongoing Phase 3), PT-141 (n=1,267 RECONNECT), MK-677 (n=65 two-year crossover) — the sample sizes are not small, the trials were placebo-controlled, the primary endpoints were prespecified, and the placebo response was measured directly and subtracted from the active-arm effect. The "it's just placebo" claim against any of these specific molecules is not consistent with the actual trial design. For the peptides where the human evidence is small or absent — the gray-market healing class (BPC-157, TB-500, KPV), the Russian cognitive trio (Selank, Semax, Epitalon), the longevity/mitochondrial class (MOTS-c, SS-31 outside Barth syndrome), Dihexa — the placebo concern is more pertinent. Most user-reported benefit from these molecules in self-experimentation contexts is unblinded, uncontrolled, and on outcomes that are particularly susceptible to expectancy. Treating those reports as equivalent to RCT evidence is exactly the error the objection criticizes, and the peptide pages on those molecules say so explicitly in their first paragraphs. The right framing: placebo concerns matter most where the trial design does not address them — small studies, unblinded use, subjective endpoints, regression to the mean. They matter much less where the trial design specifically addressed them — large randomized double-blind placebo-controlled studies with prespecified objective endpoints. Applying the placebo critique uniformly across the corpus drags the well-supported peptides down to the level of the under-supported ones rather than separating the two. Where the critic has a real point: anyone basing an applied use of BPC-157 (or Selank, or Dihexa) on personal experiment is acting on an evidence base where the placebo concern is genuinely operative, and confidence in the molecule's effects should be calibrated accordingly. The marketing layer of the peptide field consistently overstates the certainty available from small studies, and the "it's just placebo" critique is the corrective rhetoric that pushes back on that. Where the critic loses the thread: extending the legitimate critique of small unblinded trials to the FDA Phase III evidence base for Tesamorelin, Semaglutide, Tirzepatide, or PT-141 produces a sentence that is uninformed about the trial designs it is criticizing. Placebo-controlled means the placebo response was measured. The active-arm effect is what is left after that subtraction.