Your HRV baseline Baseline Your own usual HRV level and range, built from weeks of readings taken the same way. Comparing a reading with your baseline says more than comparing it with other people. More about Baseline is your own typical range, built from a few weeks of readings taken the same way. Most people need two to three weeks of consistent readings to build one. After that, compare your 7-day average with your usual range rather than reacting to single days.
This is the last lesson of Level 2. It uses the 7-day log you started in lesson 9.
What a baseline is, and what it isn’t
A baseline has two parts: your typical value, and how far you normally move above and below it from day to day. Together they describe what “normal for you” looks like.
A baseline is a reference point for you alone. A baseline of 35 ms is no worse than someone else’s 70 ms, for the reasons in lessons 4 and 7. It also drifts slowly over the years, which is why apps keep updating it.
For daily use, your baseline beats any population table. A norm table compares you with strangers measured in a lab. Your baseline compares you with yourself, measured the same way, with the same device.
How many days you need
Expect to need two to three weeks. Garmin’s HRV Status, for example, needs three weeks of consistent sleep data before it shows a status at all.
The reason is noise. HRV moves a lot from one day to the next, even when nothing in your life has changed. You need enough days to see both your average and your usual spread. In the triathlete study described below, each athlete’s normal range came from their first two weeks of readings (Plews and colleagues, 2012).
Do not confuse this with recording length. Studies that say “a minute is enough” are about how long each single reading lasts, not how many days you need.
Use averages, not single days
Single days are noisy. Buchheit’s 2014 review puts the typical day-to-day variation of resting lnRMSSD lnRMSSD The natural logarithm of RMSSD. Taking the log evens out the spread of RMSSD values, which makes day-to-day changes easier to compare and analyze. More about lnRMSSD at about 12%. So a one-day drop of 10% can be ordinary noise.
The answer from sports science is the rolling average: each day, average the last 7 days. In elite endurance athletes, HRV averaged over a week tracked 10-km running performance, while single-day values did not (Plews and colleagues, 2013). Buchheit’s review (2014) likewise reports that links with performance only showed up when at least 3 to 4 days were averaged. Garmin’s HRV Status works from a seven-day average too.
Here is an example week of RMSSD RMSSD Root mean square of successive differences: an HRV metric built from how much each heartbeat interval differs from the next. It mainly reflects vagal activity. More about RMSSD readings:
| Day | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
|---|---|---|---|---|---|---|---|
| RMSSD (ms) | 52 | 48 | 55 | 41 | 50 | 54 | 49 |
| lnRMSSD | 3.95 | 3.87 | 4.01 | 3.71 | 3.91 | 3.99 | 3.89 |
The 7-day average RMSSD is 349 ÷ 7 ≈ 49.9 ms (lnRMSSD ≈ 3.91). Day 4 looks like a big drop on its own, but it moves the 7-day average by less than 2 ms. On day 8, drop day 1 and add the new reading. If day 8 is 58 ms, the new average is 50.7 ms.
Go deeper Why sports scientists average the log
Researchers usually average lnRMSSD rather than raw RMSSD, for the reasons in lesson 6: the log evens out the skew, so a few very high days do not pull the average up too much. If your app gives you RMSSD, you can average RMSSD at home. Keep using the same one.
To turn an lnRMSSD average back into milliseconds, use e to that power: e^3.91 ≈ 50 ms.
Your normal range
Your normal range is the band around your baseline that your 7-day average usually stays inside. A change starts to matter when it takes you outside that band and keeps you there.
Sports scientists usually build the band from your own data: your average over the first few weeks, plus or minus a fraction of your day-to-day spread. Studies have used either half or all of an athlete’s own day-to-day variation as the line for a worthwhile change. Buchheit (2014) points out that there is no evidence yet that any particular fraction is the right one for practical decisions.
An older study used a similar idea to guide training. Men in the HRV group trained hard when their morning HRV was normal, and easy or not at all when it fell below their own 10-day average minus one standard deviation. In this small study of 26 men, their fitness improved over four weeks (Kiviniemi and colleagues, 2007).
Apps build their own bands, and the exact methods are mostly not published. Garmin calls your status “balanced” when your seven-day average is inside your baseline range, “unbalanced” when it is above or below, and “low” when it is well below (see what unbalanced HRV status means). No universal cut-off, such as “a 10% drop”, works for everyone.
How much your HRV bounces: the coefficient of variation
The coefficient of variation (CV) is your day-to-day spread divided by your average, given as a percentage. It tells you how much your HRV bounces around.
In the example week above, the CV of RMSSD is about 9%.
Researchers also track how the CV changes over time. In a case study of two elite triathletes, the athlete who became overreached showed a steady fall in the 7-day lnRMSSD average and in its CV over the weeks before the race. The other athlete’s values stayed stable (Plews and colleagues, 2012). That is one athlete, so treat the CV as a detail to watch, not a rule.
Four common patterns
| Pattern | What it looks like | Possible reasons | What to do |
|---|---|---|---|
| Stable | 7-day average stays inside your range | Normal life | Nothing; keep measuring the same way |
| Dip and rebound | A drop for a day or two, then back to normal | Alcohol, a late night, a hard session, travel | Note the cause; ease off if you feel run down |
| Slow decline | The 7-day average drifts down for one to two weeks | Building stress, too much training, the start of an illness | Look for causes; rest more; watch for symptoms |
| Sustained rise | The 7-day average sits above your range for a week or more | Recovery or rising fitness, but also poor readings or heart rhythm changes | Check your readings are clean; read lesson 4 again |
A rise is not always good news, as lesson 4 explained. For more on day-to-day swings, read why your HRV changes so much, and for training decisions, should you train when your HRV is low?
Add context: tag your days
Numbers make more sense with notes. In 28,175 people who measured each morning with an app, RMSSD was on average 12% lower after heavy drinking and 10% lower when they were ill, compared with their own usual values (Altini and Plews, 2021). Tag each day with anything that might matter: alcohol, a late meal, a hard workout, poor sleep, illness, travel, and your cycle phase if you menstruate (see HRV and the menstrual cycle). After a few weeks, patterns appear. Change one thing at a time if you want to know what really moves your HRV. The what affects HRV page covers the main factors.
When to act, and when to see a doctor
HRV alone never diagnoses anything. But a lasting drop can be one sign among others that your body is under strain. Talk to a doctor if a change in your HRV comes with symptoms such as fever, chest pain, palpitations, shortness of breath, fainting or unusual tiredness that does not go away. If you feel well and the dip rebounds, it is most likely ordinary noise.
You have finished Level 2. Level 3 is about using HRV well. It starts with what moves HRV.