Almost every supplement review you'll find online is written by someone who thinks it worked.
That's not a coincidence. It's a selection effect embedded deep in how supplement culture produces content. The person who tried lion's mane for three months and noticed nothing doesn't write a blog post. The person who noticed sharper focus in week two does. The result: an ecosystem of testimonials systematically biased toward positive outcomes.
The uncomfortable truth this creates: you probably have items in your supplement stack that aren't doing anything for you. And you have no reliable way to know which ones — because every signal you'd rely on (feeling better, performing better, looking better) is subject to the same biases that produce misleading reviews.
This is where the anti-placebo concept becomes one of the most useful ideas in self-experimentation.
What Is the Anti-Placebo?
The placebo effect is the measurable improvement that occurs when you believe a treatment is working, even when it contains no active ingredient. It's real, it's meaningful, and it's the reason clinical trials have control groups.
The anti-placebo is the inverse phenomenon: discovering, through rigorous self-testing, that a supplement you expected to work doesn't move your personal metrics. A negative result.
The supplement industry is built on the assumption that people want to hear "this works." But for a data-driven experimenter, a negative result is information. It's not a failure — it's a finding. And financially, it's often worth more than a positive result.
In formal science, negative results are considered equally valid as positive results. The null hypothesis (this does not have the expected effect) is just as informative as its alternative. The bias against publishing negative results — known as publication bias — is one of the most studied problems in research methodology.
Related: Want to put this into practice? Try our Supplement Stack Audit to get started, and check out Biohacking on a Budget: Best Interventions Under $50 for more context.
Why Self-Experimentation Is Unreliable Without Controls
The challenge of evaluating supplements without a rigorous framework is that the human brain is an exceptional pattern-recognition machine that is also deeply prone to seeing patterns that aren't there.
Confirmation bias: You read that ashwagandha reduces anxiety, you buy it, you notice yourself feeling calmer two weeks later. Did the ashwagandha cause that? Or did the act of doing something about your stress (buying, taking, believing) create that shift? Did circumstances change? Did you sleep better randomly? Did you reduce caffeine without noticing?
Regression to the mean: People often start supplements when they're feeling bad. They would have felt better anyway — conditions fluctuate — but the supplement gets the credit.
Nocebo in reverse: When you're paying $80/month for something, you're motivated to perceive benefit. The investment creates a cognitive pressure to feel like it's working.
Single variable contamination: Most people change multiple things at once. They start NMN the same week they download a sleep optimization app, clean up their diet slightly, and cut back on alcohol. Something improved. What was it?
How to Design a Fair Individual Trial
A rigorous personal trial doesn't require a control group of identical twins. It requires a structure that separates signal from noise.
Establish a Baseline First
Run at least three to four weeks of consistent tracking before adding anything new. Measure your target outcomes daily or weekly: HRV, sleep score, subjective energy (1–10), perceived stress (1–10), training performance, cognitive clarity — whatever the supplement claims to affect.
This baseline is critical. Without it, you cannot detect change. Many people start tracking after starting a supplement, which tells them nothing about whether they actually improved.
Change One Variable
Do not change anything else during the test period. Same diet, same sleep routine, same training volume, same other supplements. If you're introducing a new supplement, only introduce that one.
This sounds obvious. Most people don't do it.
Define Your Outcome Metric in Advance
What specific thing would need to change for you to conclude the supplement is working? This question should have a concrete answer before you open the bottle.
- "My seven-day average HRV should increase by at least 5ms"
- "My sleep score should improve by at least 4 points"
- "My perceived stress should drop by at least 2 points on a 10-point scale"
Defining this in advance prevents the outcome from being redefined after the fact to match whatever actually happened.
Run Long Enough
Different compounds have different timelines. Creatine takes two to four weeks to saturate. Vitamin D takes six to eight weeks to shift serum levels. Ashwagandha shows cortisol effects around weeks four to eight. Running a trial for two weeks and drawing conclusions is usually not enough signal.
A minimum viable personal trial: four-week baseline tracking, eight weeks on the supplement, four weeks post-discontinuation to assess whether metrics change back. Twelve to sixteen weeks total. This catches slow-onset effects and confirms whether any change was durable.
Common Supplements That Don't Move the Needle for Many People
These aren't universally ineffective — the science on some is genuinely interesting. But the individual non-response rate is high enough that rigorous personal testing is particularly warranted.
NMN and NR: NAD+ precursors have produced compelling results in animal models and some early human pharmacokinetics research. Affordable oral doses (250–500mg) do raise blood NAD+ levels. Whether that translates to measurable outcomes in healthy younger adults — energy, cognition, body composition, longevity markers — is far less clear. Many users who track carefully report no detectable change in any metric after three months.
Collagen peptides for joint support: Some evidence for reducing joint discomfort in people with existing knee pain. For younger, healthy individuals, the signal is weak. Very expensive relative to protein sources that also provide the same amino acids.
Branched-chain amino acids (BCAAs): In people who eat adequate protein, BCAAs add essentially nothing. The muscle protein synthesis benefit is driven by leucine threshold, which you hit easily if you consume sufficient total protein. Many gym-goers take expensive BCAA powders on top of already-sufficient protein intake.
Testosterone "boosters" without clinical-grade ingredients: Products with vague "proprietary blends" and no third-party testing tend to show no measurable testosterone change in bloodwork. Zinc and vitamin D may genuinely support testosterone in deficient individuals — but those require targeted supplementation, not proprietary blends.
Many nootropic blends: The evidence behind individual branded nootropic stacks is almost always inferior to the evidence for their constituent ingredients tested individually. Blends are a way to make dosing opaque and ingredient evidence claims indirect.
Be the first to try Prova
We're building an app to track whether self-experimentation and supplement tracking actually works. Join the waitlist.
The Financial Case for Negative Results
The value of discovering that something doesn't work for you is concrete and calculable.
If NMN at $80/month isn't moving any of your tracked metrics, that's $960/year you could redirect to something that does move them. If a collagen supplement at $25/month produces no joint benefit you can detect, that's $300/year saved. Across a full stack with five non-responder items at an average of $30/month each, that's $1,800/year.
Over three years, the financial value of rigorous self-testing that identifies non-responses exceeds $5,000 in a typical premium supplement stack.
That money could fund annual bloodwork, a quality wearable device, a gym membership, or meaningful investment in dietary quality — all of which have stronger evidence for health outcomes than most premium supplements.
Pros
- +Negative results directly save money — often hundreds of dollars per year
- +Simplifying your stack makes causal attribution clearer for everything remaining
- +Builds genuine self-knowledge rather than reliance on population-level claims
- +Forces a baseline tracking practice that improves overall health awareness
- +Aligns with scientific epistemology — null hypotheses matter
Cons
- -Requires discipline to maintain single-variable isolation
- -Some benefits (longevity, cancer prevention) are unmeasurable in personal trials
- -Trial duration requirements are long — several months per compound
- -Placebo effect is real and positive — eliminating it has a cost
- -You may discover you've been spending money on things that didn't work, which is uncomfortable
Why Negative Results Are Scientifically More Valuable
In formal research, a null result under rigorous conditions is considered stronger evidence than a positive result under weak conditions. A well-controlled trial showing no effect is more informative than an uncontrolled trial showing an effect — because the positive result could be noise, while the negative result under control conditions suggests the effect genuinely isn't there.
Your personal trials operate by the same logic. A sloppy experiment that "shows" a supplement works tells you nothing. A well-structured eight-week trial with a defined baseline, controlled variables, and pre-specified outcomes — that shows no change — is genuine information about your biology.
The supplement industry is built on selling you hope. The anti-placebo mentality is built on something more valuable: actual knowledge about what works for you specifically.
That's worth more than any supplement on the market.
The fastest way to build a personal evidence base is to track consistently before, during, and after each supplement trial. The key metrics: HRV (morning average), sleep score, and two to three subjective ratings that match your supplement goals. Four weeks of baseline data before any trial begins.