What Actually Moves My HRV?

You have been watching the number for a few months. Some mornings it is 30 points down for no reason you can name. You changed nothing, or you changed four things at once, and the app offers no ranking of which one mattered. So the number becomes something to react to rather than something to read.

The useful reframe is that heart rate variability is not one input with one cause. It is a summary of how much your autonomic nervous system is being asked to do, and the things asking are on wildly different timescales. Most of the confusion comes from comparing a fast input to a slow one and expecting them to behave the same way.

The number is a nightly average of a moving target

Before ranking anything, it helps to know what your app is actually reporting. Almost all consumer devices take HRV during sleep, and most report either an average across the night or a reading from a stable window in the early hours.

That matters because HRV moves constantly. It rises and falls with each breath. It differs between lying down and sitting up. It changes across the night as you move through sleep stages. A single daytime spot reading and a nightly average are different measurements, and neither is wrong; they answer different questions.

So when a number moves, the first question is not "what did I do" but "is this the same measurement as yesterday". If you changed devices, changed which wrist or finger, or started sleeping in a much warmer room, you may be comparing two different things. Our guide on why Oura and WHOOP give different scores covers how far apart two devices can sit on the same night.

What moves it a lot, and fast

These act within hours and usually show up the same night or the next one.

Alcohol. This is consistently the largest single-night mover for most people who track it, and the effect is dose related and lasts beyond the night you drank. It is common enough that it is worth its own guide: how much alcohol affects HRV.

Immune activity. A cold coming on, a vaccination, an infection you have not noticed yet. HRV often drops before you feel unwell. This is a signal about physiological load, not a diagnosis, and it is not a substitute for medical assessment.

Late or large meals. Digestion during the measurement window raises heart rate and suppresses variability. Eating close to bed frequently produces a lower reading with no other cause.

Sleep timing, more than sleep duration. Going to bed three hours later than usual tends to hit the number harder than losing an hour. Shift work and long-haul travel are the extreme version.

Hard or unaccustomed training in the previous day or two. Note that this is normal and expected. Training load is supposed to produce a temporary suppression.

Heat. A hot room, a hot bath close to bed, a hot climate. Thermoregulation during sleep raises heart rate and lowers HRV.

Stress you are still carrying at bedtime. Not stress in general, but an argument at 10pm or a deadline you are still turning over.

What moves it slowly, over months

These are the ones people most want to change and most often measure over too short a window.

Aerobic fitness. Consistent training volume built over months tends to raise resting HRV, though the size of that effect varies a great deal between people. Age, which trends downward across the lifespan and is the reason a population range answers almost nothing about you personally. Body composition. Chronic sleep debt as a sustained pattern rather than one bad week. Long-running psychological stress, which behaves like a slow input even though it feels like a fast one.

The mismatch here is the whole problem. If you start training to raise your HRV, the first thing that happens is that your HRV goes down, because the training load is a fast input and the fitness adaptation is a slow one. People routinely abandon a working programme in week two on the strength of a signal that was telling them it was working.

What gets blamed and mostly does not move it

A single supplement, taken for a week, evaluated against a noisy nightly number. The effect size, if there is one, is usually far smaller than the night-to-night variation, which means a week of data cannot separate them. This is the general problem covered in how to tell if a supplement is working.

One good night after a bad week. One cold shower. A breathing session done once. Breathing protocols do raise HRV measurably while you are doing them, which is not the same as raising your overnight baseline.

None of these are useless. They are simply being asked to prove themselves against a measurement too noisy to see them at this sample size.

The ranking that matters is not this one

Everything above is the general pattern. The order for you is personal, and it is discoverable.

Two people can have opposite dominant inputs. For one, alcohol swamps everything and sleep timing barely registers. For another, a 30 minute shift in bedtime moves the number more than two drinks. A published ranking cannot tell you which you are. Your own data can, but only if you stop changing four things at once.

This is also why a personal reference beats a population one. A number that is unremarkable for you can sit well outside a published range, and the reverse. That argument is made in full in what a personal health baseline is.

What to do with it tomorrow morning

Read direction, not the daily value. A single morning inside a normal amount of scatter carries very little. Four weeks of direction carries a great deal.

Change one thing at a time, and give it long enough to show. Fast inputs need days. Slow inputs need a couple of months. Testing a slow input over a week produces a confident answer that is not connected to reality.

Expect a low reading after a hard session, and do not treat it as a failure. If you feel fine and the number is down, that combination is common and is covered in HRV dropped but I feel fine.

Stop comparing yourself to other people's numbers. The single most common source of unnecessary worry here is a published average. There is more on that in what a good HRV is for my age.

The number is worth watching. It is just answering a narrower question than most apps imply, and it answers it best when you already know which input you are asking about.