Study S4 · August 2026
Body Battery, HRV and sleep score: three metrics or one?
Key finding
Body Battery moved with HRV at a median within-athlete correlation of 0.674 and with sleep score at 0.683. Every athlete was positive on both, and 98% cleared 0.3. Checking all three each morning is largely reading one signal three times.
Gneta, August 2026 · 82–93 athletes · 30+ paired days each
0.674
Body Battery ↔ HRV
0.683
Body Battery ↔ sleep
100%
of athletes positive
What we did
For every athlete with at least 30 paired days, we computed the Spearman rank correlation between their morning Body Battery and their overnight HRV, and separately between Body Battery and sleep score. Cohort sizes differ per pair because not every athlete has every metric recorded — 82 for HRV, 85 for sleep score, 93 for the stress control.
We then ran two controls with known physiological answers, to check the pipeline could detect a relationship at all.
What we found
Finding 1
The three recovery numbers move together, for essentially everyone
Body Battery against HRV: median 0.674 (IQR 0.574–0.793) across 82 athletes. 100% positive, and 97.6% above 0.3.
Body Battery against sleep score: median 0.683 (IQR 0.618–0.760) across 85 athletes. Again 100% positive, and 98.8% above 0.3.
Unanimity across 82 and 85 independent athletes is not a subtle statistical signal. Whatever these three numbers are built from, they share most of it.
Finding 2
Both controls landed exactly where physiology says they should
HRV against resting heart rate — a genuinely inverse pair — came out at a median of −0.696 across 82 athletes, with 95.1% below −0.3 and not one athlete positive.
Body Battery against stress came out at a median of −0.452 across 93 athletes.
Controls that behave correctly are what let you trust the headline number. Ours recovered a known inverse relationship at the right sign and roughly the right strength.
Finding 3
What this does and does not mean
A high correlation between two scores does not tell you either one is accurate. It tells you they carry overlapping information. All three could be right together, or wrong together, and this study cannot separate those.
What it does support is narrower and still useful: if you already look at HRV, the additional decision-relevant information in Body Battery is smaller than having three separate dials implies.
Limits you should know before citing this
- The cohort is self-selected. Garmin owners who sought out a third-party analytics tool. Fitter and more data-curious than average, and not a random sample of anyone.
- Correlation is not accuracy. Nothing here validates any of the three against an external standard. They agree with each other; that is all.
- Shared inputs are expected, not scandalous. Garmin has never claimed these are independent measurements. Body Battery is documented as drawing on HRV, stress and sleep. This study measures how much that overlap shows up in practice.
- Cohort sizes differ between pairs. 82, 85 and 93 athletes for the three comparisons, because coverage of each metric varies. The pairs are not strictly like-for-like subsets.
- Six accounts excluded. Near-identical multi-year histories under different Garmin accounts, created inside a known incident window. See the methodology.
- Observational. No causal claim is made anywhere in this study.
The data
The aggregate output and the script that produced it are public. Every number on this page traces to that file.
View the aggregate data (JSON) →