Daily HRV trends and their coefficient of variation: athletes with lower fitness or higher perceived fatigue show larger lnRMSSD reductions under high training load
Participants
n=10 collegiate female soccer players.
Protocol
Ultra-short, smartphone-derived lnRMSSD (log-transformed root mean square of successive R-R differences) and psychometrics recorded daily throughout 2 consecutive weeks: one high training-load (TL) week and one low TL week
Fitness parameters assessed after the training period
Outcomes: team and individual daily lnRMSSD trends vs baseline; lnRMSSD fluctuation expressed as the coefficient of variation (lnRMSSDcv) and its relationship to perceived fatigue and fitness
Key Findings
Training load differed strongly between the two weeks (ES ± 90% CL = 1.37 ± 0.80)
High TL week: lnRMSSD reductions vs baseline ranged from unclear to very likely moderate (ES = -0.21 ± 0.74 to -0.64 ± 0.78)
Low TL week: lnRMSSD reductions were unclear (ES = -0.03 ± 0.73 to -0.35 ± 0.75)
Higher lnRMSSDcv (more day-to-day fluctuation) was associated with greater perceived fatigue and lower fitness (r = -0.55 [-0.84, -0.003] and -0.65 [-0.89, -0.15], both large)
Authors' conclusion: athletes with lower fitness or higher perceived fatigue showed greater lnRMSSD reductions throughout training — useful when interpreting individual lnRMSSD responses for managing fatigue
Practical Application
In 10 collegiate female soccer players, the day-to-day fluctuation of HRV (lnRMSSD coefficient of variation) tracked perceived fatigue and fitness: more fluctuation went with more fatigue and lower fitness, and lnRMSSD fell more during a high-load week than a low-load week. Read HRV as a trend and as a spread, not as a single morning number.
Not from this paper; the base's own practice: use a rolling weekly mean of lnRMSSD as the primary monitor and plot its weekly CV alongside; a rising CV with a falling mean is the pattern to treat as a load-reduction signal, a stable mean with a stable CV as coping, a rising mean with a stable CV as positive adaptation.
Not from this paper; the base's own practice: do not react to a single-day drop unless subjective symptoms or a sustained trend accompany it; trend analysis needs regular measurement, so aim for most mornings of the week.
Inference, not tested: the fitter athlete is the more stable one on HRV — a reason to expect individual, not team-wide, thresholds.
Limitations
n=10, one team, one sex, two weeks — the correlations are large but the sample is tiny and the 90% limits are wide (one reaches -0.003)
Ultra-short smartphone recordings; fitness assessed only after the training period
The rolling-mean lookback and the "4-5 measurements per week" rule that this entry used to carry are house practice, not findings of this paper
How this entry was checked and changed
Why this level
regraded 2026-09-27 from B+: small observational cohort, n=10, two weeks of monitoring
Content correction 2026-09-27
"Day-to-day HRV values are noisy (CV ~10-20% within individual)" -> not in the abstract; removed.
The rolling 7-day mean, the mean/CV decision table, the "4-5 valid measurements per week" rule and the app recommendations -> not from this paper; kept only where useful and labelled as the base's own practice, apps removed.
Level B+ -> B (observational, n=10, two weeks); Tier 4-5 (Elite) -> Tier 2-3 (collegiate); title rewritten to the study's actual finding.