Methodology
This is a public commitment. Every protocol verdict on this site passes two gates before it goes live. If it hasn't cleared both, it's marked "Still testing."
The science gate
≥2-3 peer-reviewed studies pointing in the same direction. One study is a signal worth tracking, not a conclusion. Effect size has to be statistically significant AND practically meaningful for athletes training at high intensity — not just lab results on sedentary subjects.
If primary research funding comes from the brand whose product is being tested, I flag it explicitly in the article.
The personal gate
≥3-4 weeks of personal testing across at least one full training microcycle. I track Garmin data before, during, and after: HRV status, Body Battery, sleep score at minimum. Same conditions, same timing, same execution. Random usage doesn't count.
The verdict scale
- Useful
- Science and 3-4 weeks of personal data point in the same direction.
- Mixed
- Science says yes but my data doesn't confirm it, or the reverse.
- Overhyped
- Marketing claim not supported by the research, or contradicted by my personal experience.
- Still testing
- Published with an explicit promise to update with my data at a specific future date.
What I don't write
AI does not write my verdicts, my data interpretations, my terrain experience, my contrarian opinions, or my IG captions. AI aggregates research and structures drafts. I fill in the truth.