Two numbers, a slash, and a surprising amount of anxiety. Blood pressure readings are simpler than they look, and far less dramatic than one alarming number makes them feel.
Here is what each number means, why the way you sit changes the result more than most people realize, and how to get readings worth paying attention to.
Key takeaways
The top number is the push when your heart beats. The bottom is the pressure between beats.
One reading proves very little. Among adults whose first reading landed in a high range, 35% averaged below that range once three readings were taken (Muntner et al. 2019, American Heart Association).
How you sit matters. An unsupported back can add 5 to 15 points to the top number (Muntner et al. 2019).
Smartwatches and other cuffless devices are not recommended for measuring blood pressure (2025 AHA/ACC guideline).
What do the two blood pressure numbers mean?
Blood pressure is the force of blood pushing against your artery walls, measured at two moments in every heartbeat.
The top number, systolic, is the pressure at the instant your heart squeezes and pushes blood out. The bottom number, diastolic, is the pressure between beats, while your heart relaxes and refills (American Heart Association).
Picture water in a garden hose: a surge when the pump fires, a lower resting pressure in between. That is your two numbers. Your cuff labels them mmHg, a unit of pressure. We will just call them points.
Both matter, though not equally. The top number is more strongly linked to cardiovascular risk, particularly as people get older, and in some studies, once researchers accounted for the top number, the bottom number added almost nothing extra (Muntner et al. 2019).
What do the blood pressure categories mean?
Health organizations sort readings into four bands:
Category | Top number | and / or | Bottom number |
|---|---|---|---|
Normal | under 120 | and | under 80 |
Elevated | 120 to 129 | and | under 80 |
High, stage 1 | 130 to 139 | or | 80 to 89 |
High, stage 2 | 140 or higher | or | 90 or higher |
Source: American Heart Association
Three things worth knowing about that table.
If your two numbers fall in different rows, the higher one applies. A reading of 125/85 counts as stage 1, driven by the bottom number.
A category is not a diagnosis. The AHA is explicit that these bands are meant to be applied to an average of at least two readings taken on at least two separate occasions, not to whatever your cuff said this morning.
The categories are current. A new blood pressure guideline was published in August 2025, and it kept these four bands unchanged. What it did change was how clinicians assess risk and when they start treatment, along with a stronger emphasis on readings taken outside the doctor's office (Jones et al. 2025 AHA/ACC guideline; plain-language ACC summary).
Why isn't one reading a verdict?
Because blood pressure moves constantly, and the research on this is striking.
Look at what happens when you simply measure more than once. Among US adults whose first office reading fell between 140 and 159 over 90 to 99, 35% averaged below 140/90 once the mean of three readings was used. Going the other direction almost never happened: only 3% of people who looked fine on a single reading crossed the threshold on a three-reading average (Muntner et al. 2019).
Read that again, because it is the most useful fact on this page. A third of people who get an alarming first reading are not actually in that range. They were just measured once.
The same statement notes you can never get the exact same number twice, because pressure changes from one heartbeat to the next. Your pressure is also supposed to fall while you sleep, typically by 10% or more, and rise again in the morning.
That is why guidelines ask for averages across occasions, and why a screening body recommends confirming readings outside the clinic before anyone starts treatment (US Preventive Services Task Force, 2021). It is also the whole reason Knit holds a range over time rather than reacting to any single entry, the same way we compare you to you across every other measure.
How do you take a home reading you can trust?
This section is worth more than any other on this page, because bad technique does not produce slightly-off readings. It produces readings wrong enough to change which category you land in.
How you sit changes the number:
What you do | Effect on the top number |
|---|---|
Back unsupported | adds 5 to 15 |
Legs crossed | adds 5 to 8 |
Arm resting in your lap | adds about 4 |
Arm hanging at your side | adds about 6.5 |
Sources: Muntner et al. 2019; arm position figures from a 2024 randomized crossover trial of 133 adults (Liu et al. 2024, JAMA Internal Medicine)
A review pooling 328 studies found that individual measurement errors could shift a reading anywhere from 24 points below to 33 points above a person's true resting pressure (Kallioinen et al. 2017). That is the difference between "normal" and "call your doctor," created entirely by how someone sat.
The full checklist, from the American Heart Association's home monitoring guidance:
Empty your bladder first.
Sit quietly for at least 5 minutes before the first reading.
Back supported, feet flat on the floor, legs uncrossed.
Rest your arm on a table so the cuff sits at heart level. Do not hold your arm up.
Put the cuff on a bare arm, directly above the bend of your elbow.
No caffeine, exercise, or smoking for 30 minutes beforehand.
Do not talk or use your phone during the reading.
Check the cuff fits: it should wrap around most of your upper arm. A cuff that is too small for the arm is the single most common measurement error (Muntner et al. 2019).
Take two readings a minute apart, at the same time each day.
One clarification on the bare-arm rule, since a widely repeated claim gets this wrong. You will read that clothing adds enormous amounts to a reading. When researchers actually tested it on 162 people, measuring over a sleeve versus a bare arm differed by 0.3 points, a gap small enough to be chance (Ozone et al. 2016, Journal of Clinical Hypertension). The real reason for a bare arm is that rolling a sleeve up creates a tourniquet, and bunched fabric stops the cuff fitting properly.
For a picture worth showing a clinician, the AHA suggests two readings in the morning and two in the evening, ideally for seven days.
Why is my blood pressure higher at the doctor's office?
Because that is extremely common and it has a name.
Among people whose office readings land in the high range, 15% to 30% have what is called white coat hypertension: pressure that runs higher in a clinical setting than it does in daily life (Muntner et al. 2019). Among people already on multiple blood pressure medications with high office readings, 30% to 40% show the effect.
The reverse also exists, and it is called masked hypertension. Roughly 15% to 30% of adults with normal office readings have higher pressure outside the clinic, which matters more because it goes unnoticed.
Neither pattern means the office reading is fake or that home readings are the only truth. It means one setting alone gives an incomplete picture, which is exactly why home readings have become central to how blood pressure is assessed. If your readings run consistently higher in clinic than at home, the white coat effect is worth understanding in more detail.
Useful conversion when comparing: an office reading of 140/90 corresponds roughly to 135/85 at home, since home readings tend to run slightly lower (Muntner et al. 2019).
What actually lowers blood pressure?
Diet and movement have real, measured effects. Two of the best trials give specific numbers, and it is worth being precise that these are averages from controlled studies, not promises about any individual.
The original DASH trial randomly assigned 459 people to one of three diets for eight weeks. Those eating the DASH pattern, rich in fruits, vegetables and low-fat dairy, ended up with a top number 5.5 points lower on average than those on a typical American control diet. Among participants who started with high blood pressure, the drop averaged 11.4 points (Appel et al. 1997, New England Journal of Medicine).
Cutting sodium adds to that. In a follow-up trial, cutting sodium from a high to a moderate intake lowered the top number by about 2 points, and cutting it further to a low intake lowered it another 4.6 (Sacks et al. 2001). Combining a DASH-style diet with low sodium, versus the control diet at high sodium, produced readings about 7 points lower in people without high blood pressure and 11.5 points lower in those with it.
Regular physical activity is part of the same guideline advice, alongside limiting alcohol and losing weight if you carry extra (American College of Cardiology summary). If you want the least intimidating place to start, daily step count is it.
A note on smartwatches
Worth saying plainly, because it affects how you should read anything a wearable tells you about blood pressure: the 2025 guideline explicitly does not recommend cuffless devices, including smartwatches, for diagnosing or managing high blood pressure (American College of Cardiology summary).
A validated arm cuff is still the tool, meaning one tested against measurement standards rather than simply sold as a blood pressure monitor. What an app like Knit is genuinely good for is holding the readings you take with that cuff, turning scattered numbers into a range you can see over months, and making them easy to bring to an appointment.
When should you seek help right away?
If your top number is 180 or higher, or your bottom number is 120 or higher, wait a few minutes and measure again. If it stays that high, contact your clinician promptly.
If a reading in that range comes with chest pain, shortness of breath, back pain, numbness or weakness, vision changes, or difficulty speaking, call 911. That combination is an emergency, not something to monitor at home (American Heart Association).
Frequently asked questions
What is a normal blood pressure?
Under 120 for the top number and under 80 for the bottom, according to the American Heart Association. Readings of 120 to 129 over under 80 are labeled elevated, and 130 or higher, or 80 or higher on the bottom number, falls into the high ranges (AHA). These bands are meant to be applied to an average of at least two readings on at least two occasions, not a single measurement.
Which number matters more, the top or the bottom?
Both are used, but the top number carries more weight for cardiovascular risk, especially with age. In some research the bottom number showed little independent association once the top number was accounted for (Muntner et al. 2019). If your two numbers land in different categories, the higher category applies.
Why is my blood pressure different every time I check?
Because it genuinely changes minute to minute, and because small differences in how you sit have large effects. An unsupported back can add 5 to 15 points, crossed legs another 5 to 8, and an arm hanging at your side about 6.5 (Muntner et al. 2019; Liu et al. 2024). This is why averages across several readings mean far more than any single number.
Can my smartwatch measure blood pressure?
Not reliably enough to be used for it. The 2025 blood pressure guideline specifically does not recommend cuffless devices, including smartwatches, for diagnosing or managing high blood pressure (American College of Cardiology). Use a validated arm cuff, and use apps to track and organize what the cuff tells you.
How often should I check my blood pressure at home?
For a picture worth sharing with a clinician, the AHA suggests two readings a minute apart in the morning and two in the evening, ideally across seven days (AHA home monitoring guidance). Outside of a focused week like that, a consistent routine matters more than frequency.
One thing home readings are never for: changing or stopping a prescribed medication. That decision belongs with the clinician who prescribed it.
Knit shows patterns, not diagnoses. Talk to a clinician about readings that concern you.
