For Athletes
Training & analysis
Sleep: what to log and what good looks like
How sleep hours and sleep quality interact, and what trend patterns matter.
What Grepiac captures
The WellnessLog model has two sleep fields:
sleep_hours (decimal, e.g. 7.5): hours slept last
night.
sleep_quality (1 to 10): subjective quality, 1 =
terrible, 10 = perfect.
Both are optional, but having both produces a clearer
signal than either alone.
Why two fields
Hours alone misses the case where you were in bed 8 hours
but slept poorly. Quality alone misses the case where you
slept great for 5 hours but were short. Together:
- Good hours + good quality: rested.
- Good hours + low quality: probably stressed, or
woke up several times.
- Low hours + good quality: short but deep; OK
occasionally, problematic if recurring.
- Low hours + low quality: red flag, especially across
multiple days.
What "good" looks like for endurance athletes
- Hours: most athletes need 7.5 to 9 hours in active
training blocks; less is possible in light weeks. Below
6.5 consistently impairs adaptation.
- Quality: 7 to 9 most nights. Anything below 5 for
multiple nights is a signal.
- Variability: weekly average matters more than any
single night.
Grepiac does not prescribe a per-athlete sleep target.
Athletes vary; what stays constant is whether their own
baseline holds.
The concerning trigger
Logging sleep_hours under 5 fires a concerning flag (see
Concerning flags explained).
The flag is on the entry, not a multi-day pattern: a single
4-hour night flags.
There is no flag for low sleep quality scores alone.
Quality at 3 every night for a week reads as a clear signal
to the coach but does not auto-flag.
Patterns to watch in the trend
- Consistent loss of 30 to 60 minutes vs your typical:
often the first sign of building stress, illness, or
overreaching.
- Late-night drift (sleep onset moving later week over
week): typically work stress or screen habit, not
training.
- Multiple wake-ups (low quality, normal hours):
worth checking caffeine intake, evening exercise timing,
alcohol.
What you can do
- Be honest in logging. Smoothing the numbers does not
help.
- Add context in the notes field. "Travel", "kid sick",
"stressful work week" goes a long way to interpreting a
bad week.
- Pair with HRV and RHR if you have a device that
captures them. Multiple low-recovery signals together is
stronger than any one.
See also:
Logging daily wellness,
Concerning flags explained,
HRV: where it fits.