Placebo expectancy is the measurable shift in perceived outcome that comes from believing a supplement will help, independent of any pharmacological action. It reliably produces real changes in self-reported mood, energy, sleep quality, and pain, and those changes are strongest in the first one to two weeks after a new stack starts. Self-experimenters who judge a compound on early subjective proxies almost always overestimate its effect — sometimes by the entire apparent effect size.
The expectancy timeline
Placebo responses follow a predictable arc in most self-experimentation windows. Unfair explicitly weights review conclusions against this curve rather than treating every day's log as equivalent.
| Window | Typical placebo contribution to subjective score | What to trust |
|---|---|---|
| Day 1–3 | Very high; novelty, routine change, and anticipation dominate | Adherence only; ignore outcome claims |
| Week 1–2 | High; "honeymoon" lift on mood and energy is common | Trend direction, not magnitude |
| Week 3–4 | Moderate; expectancy fades unless reinforced | First tentative read on real effect |
| Week 4–8 | Low; sustained shift is likelier to reflect pharmacology | The decision window for most compounds |
The numbers are not exact but the shape is real. A supplement that feels remarkable on day 3 and unremarkable by week 5 is usually an expectancy effect that faded, not a compound that "stopped working."
Why expectancy is not imaginary
It is tempting to dismiss placebo as mental. The response is physical. Documented placebo effects include changes in dopamine release in Parkinson's trials, endogenous opioid release in pain studies, and measurable cortisol shifts after sham stress interventions. When a user says their mood improved after starting rhodiola, the mood change is real; the question is whether rhodiola caused it.
That question is hard to answer from a single open-label week because the user, the routine, and the expectancy all changed on the same day.
How to separate expectancy from effect
A few structural defenses work far better than willpower or self-awareness.
- Extend the trial past the expectancy window. A decision made at week 6 is substantially less confounded than one made at day 4.
- Pair subjective logs with an [objective proxy](/glossary/objective-proxy). An energy claim that shows up in both a self-score and a wearable resting-heart-rate trend is sturdier than a lone self-score.
- Establish a pre-trial [baseline](/glossary/baseline-establishment). Without a reference, the early lift looks large; against a stable 14-day baseline, it often looks like noise.
- Introduce one compound at a time. If three things changed on Monday, no log entry for the rest of the week can isolate a cause.
- Run it again later. A true effect mostly survives a second trial after a washout. A pure expectancy effect rarely repeats as strongly the second time around.
These are the same defenses the app's risk checks use on the active-compound side of a stack, applied to the interpretation side.
Expectancy vs. nocebo vs. real effect
Expectancy and its dark twin nocebo effect operate on the same mechanism but in opposite directions, and both can coexist with a real pharmacological effect. A compound can genuinely help, be subjectively oversold in week 1 by expectancy, and simultaneously produce a mild nocebo headache from reading the label. The decision framework is the same in all three cases — collect enough days past the expectancy window, require coherence across subjective and objective data, and run an n-of-1 experiment if the answer still matters after a first pass.
How this appears in Unfair
Review cycles display week 1–2 data in a muted style and only weight week 4 onward at full confidence when producing a recommendation update. The "early impression" a user records on day 3 is saved but not used to change the ranking. When an effect size on a subjective proxy collapses between week 2 and week 6, Unfair flags the drop as an expectancy-consistent pattern rather than a compound that weakened.
Clinical safety note
Placebo expectancy is a reason to be conservative about claims of benefit, not about safety. A real adverse symptom is still real even if it lands inside the expectancy window. Persistent or escalating symptoms should prompt a stack pause and clinician review regardless of where in the timeline they appear.