Blog · Recovery
Should I train today?
Your watch shows an HRV number every morning. Almost everything written about what it means skips the one question that decides it: compared to what?
The short answer
There is no good HRV number. HRV varies enormously between people — healthy adults sit anywhere from the teens to well over a hundred milliseconds — so a number that would be a red flag for one person is an ordinary Tuesday for another. The only comparison that means anything is your reading against your own recent baseline.
Read that way, the useful signal is roughly: HRV meaningfully below your own average and resting heart rate meaningfully above it, on the same morning, is worth acting on. Either one alone is a shrug. Neither one, on its own, is a reason to skip a session.
Why "what's a good HRV" is the wrong question
HRV — heart rate variability — is the variation in the gaps between heartbeats. Higher generally tracks with a more rested, parasympathetically-dominant state; lower tracks with stress load, whether that came from a hard session, a bad night, alcohol, illness, or your job.
But the absolute value is mostly a fact about you: your age, your genetics, your anatomy, and the device on your wrist. Comparing it to a stranger's is like comparing shoe sizes to decide who ran further. The day-to-day movement of your own number is the part that carries information.
This is why "is 45 ms good?" has no answer, and why the follow-up — "is 45 low for me, this week?" — has a very good one.
How many nights before the comparison is real
A baseline built from two nights is noise with a decimal point. A single bad night's sleep, a late meal, a glass of wine, or the watch sitting slightly loose can each move a reading more than a week of hard training will.
Practical rules that hold up:
- Five nights is the floor. Below that you do not have an average, you have a couple of numbers. Anything claiming to interpret them is guessing.
- Two to four weeks is where it gets useful. A rolling window of about a month is long enough to absorb one bad night and short enough to follow you as your fitness, sleep, and stress actually change.
- More readings buy more confidence, not more certainty. Twenty nights of data earns a stronger read than eight; eight earns a stronger read than five. The right response to a thin baseline is a quieter claim, not a louder one.
Yesterday's reading is not today's
A number that is three days old should not get a vote in this morning's decision. These signals describe a state, and states change overnight. If your watch was charging for two nights, the honest answer to "should I train today" is that this input has nothing to say — not that you should train on Tuesday's data.
This is the failure mode most readiness scores never show you. They keep printing a score from whatever the last reading was, and a stale number looks exactly like a fresh one.
The device trap nobody warns you about
If you have ever added a ring, a chest strap, or a sleep mattress cover alongside your watch, you have probably seen your numbers jump — and it was not your body that changed.
Different devices measure HRV in different ways, at different times of night, using different algorithms. A watch taking opportunistic samples through the day and an overnight sensor averaging your deepest sleep can produce readings tens of percent apart from the same person on the same night. Neither is lying. They are answering different questions.
The trap: if your baseline was built from one device and today's reading comes from another, every comparison is wrong, and it is wrong in a direction that flatters you. A new sensor that reads high makes every morning look "well above baseline" until the average catches up — which can take weeks, during which anything reading that signal will happily tell you to push harder.
If you switch devices, the cleanest thing you can do is treat it as a fresh start and give the new baseline its five-plus nights. Mixing the two is worse than having less data.
Reading the two signals together
Resting heart rate is the useful companion to HRV, because it usually moves the opposite way and it is far less noisy. A rough, honest way to read a morning:
- Both good — HRV at or above your average, resting heart rate at or below it. Nothing is asking you to hold back. Train as planned.
- They disagree — one flagged, one fine. This is the most common result and it mostly means "ordinary day". Look at how you actually feel, how you slept, and what you did yesterday. Do not reorganise your week around it.
- Both flagged — HRV clearly down and resting heart rate clearly up, together. This is the combination worth respecting. It does not have to mean cancelling: reducing the load, dropping the top sets, or swapping to easy aerobic work is usually a better trade than a binary train-or-rest decision.
- Both flagged for several days, alongside a raised breathing rate or temperature — that is a pattern about your health, not your programming. Rest, and if something is genuinely wrong, see a doctor rather than a dashboard.
One principle underneath all of it: the errors are not symmetrical. Backing off a day early costs you very little. Training hard through a day your body was already struggling with costs you the session, sometimes the week, occasionally an injury. When the signals are ambiguous, the cheap mistake is the one to make.
What these numbers cannot tell you
HRV does not know you had an argument, a deadline, or a four-hour flight. It does not distinguish "stressed from training" from "stressed from life", and the training response to those is not the same. It cannot diagnose anything. And it will not override the most informative signal you own, which is whether the warm-up sets felt heavy.
Used well, it is a tiebreaker on the days you are genuinely unsure, and a way to catch the weeks where you have been quietly digging a hole. Used badly, it is a reason to skip sessions you were perfectly capable of doing.
Where Senvera fits
Senvera reads these signals from Apple Health and does the comparison for you, which mostly means being careful about everything above. It builds your baseline from a rolling window of your recent nights, needs at least five readings before it will judge anything against them, and keeps baselines separate per device so a new ring is never compared against your watch's average. A reading more than a couple of days old stops getting a vote.
And when it does not know, it says so — "still learning your baseline" rather than a confident number built on three nights. That honesty is the point: a readiness score you cannot interrogate is a mood ring with a battery. Your coach then uses the result the way a good human coach would — adjusting today's demand, and telling you which signal moved and why.
Not medical advice. This article is general information for healthy adults about training decisions, not a diagnostic tool. HRV and resting heart rate are affected by many conditions and medications, and consumer wearables produce estimates, not medical measurements. If your readings change persistently, if you feel unwell, or if you have a heart condition, talk to a doctor rather than adjusting your training around a chart.