CONCEPT
The science of what works, what does not, and what is speculative.
A supplement is a compound taken to correct a gap in the diet or to produce an acute effect that food does not. In the United States, the Dietary Supplement Health and Education Act of 1994 placed them closer to food than to drugs: no premarket approval for efficacy, and the manufacturer is responsible for the claim on the label. That regulatory fact, not the biochemistry, explains most of what the industry sells and how it markets it.
Sort by how well the effect replicates in humans, at doses people take.
The default expectation for a new compound should be that it does nothing. Most do. That is not cynicism, it is the base rate.
Mechanistic plausibility is not evidence. A pathway diagram, an in vitro result, and a rodent study at a dose no human could reach are the raw material of supplement marketing precisely because they are cheap to produce and hard to refute in a sentence.
Not being a drug does not mean being inert or clean. Label accuracy is a genuine problem, and contamination with unlabeled actives has ended careers in tested sport. Third-party programs such as NSF Certified for Sport and Informed Sport exist for that reason, and their existence is the tell.
Stacking assumes effects add. Mostly they do not, because several products act on the same limited mechanism, and the body has a ceiling on the process being pushed.
Blood-panel optimization has quietly become a supplement sales channel. A marker inside the reference range is not a problem waiting to be solved, and moving one for its own sake is not an outcome.
Supplements move the last few percent of a result that training, food, and sleep produce. When those are unbuilt, nothing on the shelf substitutes for them, and the money spent looking is the clearest signal that the work has not started.