How accurate are smart watches when it comes to step counts and heart monitoring?

Danielle_B99
Danielle_B99 Posts: 33 Member
in Fitness and Exercise
Just wanted to hear your input on whether you believe the count on the watch or phone is accurate or take it with a grain of salt.

Replies

  • Lietchi
    Lietchi Posts: 6,900 Member
    It's going to depend on brand and model.

    My Garmin's step count is pretty accurate as long as I'm not on a very bumpy road :smiley: I was pleasantly surprised recently that it even counts my steps accurately when I have my hands in my coat pockets.

    Heart rate is reliable enough for day to day measurement to have a general idea, but I've had false readings sometimes: for example in very cold weather (probably can't get a good read because of constricted blood vessels) and strenuous activities, so I use a HR band (connects to my watch) for exercise, because I want to have a 'real' heart rate reading then instead of an optical one (more reliable).
  • yirara
    yirara Posts: 9,993 Member
    Optical heart rate sensors depend on a lot of things. They don't work well if you happen to have a very dark skin colour, thick skin, lots of hair, or anything else that makes the blood vessels on your wrist less visible. Wearing a device too loose can result in odd numbers as does wearing it too tight. Cold weather can influence the reading as well. And if you use the watch for exercising then sometimes cadence might show as fake heartrate if you're unlucky. But as a general trend it's overall fine.

    On thing I find very useful on my watch is heartrate variation, which is measured over night. I don't think anyone can really explain it, but I think overall a higher HR variation at night is good because your heart is ready to run away from that sabertooth tiger but also to just rest. Low variation means your body is not ready for everything. Many people reported it goes down when they drink alcohol, and I can confirm this. It dropped like a stone when I had covid, before the test was even positive, and also shows similar trends with general colds. When a subluxed tendon was put back that I ran around with for months after an accident the reading went up. I have yet to see what happens if thyroid meds are too low. But at least I find it a great indicator of something being off.
  • AnnPT77
    AnnPT77 Posts: 34,688 Member
    You asked about step counts and heart rate. If you really are wondering if they're accurate for calorie estimates, that's a very different question.

    A good brand/model should be reasonably accurate for heart rate (literally the number of beats per minute) subject to the things Lietchi mentioned that can limit their ability to get a good reading. They can also have trouble getting a reading when there's lots of arm flexion. (I have to wear a sync-ed chest belt when rowing because of that.)

    But the heart rate zones they talk about may not be accurate unless you have a sports-tested HRmax value . . . which most of us don't. If you don't input a personalized HRmax, they default to age estimates, which are inaccurate for a large fraction of people (less than a majority, but still many). Fortunately, accurate HR zones don't matter for weight loss. (The "fat burning zone" doesn't mean what many people think it means.)

    I don't have an opinion about step counts from experience (I don't care about steps), but again the good brands/models have a reasonably good reputatation.

    Calorie estimates? Well, the all-day calorie estimates will be close for quite a few people (probably the majority), noticeably high or low for a few, and quite surprisingly high or low for a very rare few. That's the nature of statistical estimates, because that's all they're giving you. They don't in any sense measure calorie burn. The issue is less the tracker's accuracy, and more about how average the particular user is in their movement patterns and such. If you follow one for a month or two, tracking food and body weight, you'll get a reasonable idea how close they are to reasonable for you.

    The calorie estimates for individual workouts . . . well, that's a more complicated question.
  • spiriteagle99
    spiriteagle99 Posts: 3,751 Member
    Step counts can be inflated if you are using your hands a lot. Or if you don't move your hands while pushing a lawn mower or stroller, it may undercount.

    HR can also be off, either because the watch is counting cadence or just because it exaggerates. My watch has shown 205 when i was 3 minutes into the run. I take it with a large grain of salt. if you are concerned that your HR seems too high, remember that the usual maximum HR formula Is an average, and some of us aren't average.
  • dougii
    dougii Posts: 679 Member
    I have found my Apple Watch Series 6 to be very accurate on both of these measurements. I've compared it's HR to multiple machines and checked it with doctor's office metrics and it is dead on or within a very small margin of error. It's step count works as well as my Garmin.
  • yirara
    yirara Posts: 9,993 Member
    dougii wrote: »
    I have found my Apple Watch Series 6 to be very accurate on both of these measurements. I've compared it's HR to multiple machines and checked it with doctor's office metrics and it is dead on or within a very small margin of error. It's step count works as well as my Garmin.

    Yes, it works for you, your skin colour/hair and your habits/activities. It doesn't mean it will work for other people as we wrote above here :)
  • I_AM_ISRAEL
    I_AM_ISRAEL Posts: 160 Member
    My Apple Watch is pretty consistent and accurate in regards to both.
  • Unknown
    Unknown
    This content has been removed.
  • yirara
    yirara Posts: 9,993 Member
    aundy28570 wrote: »
    I just purchased a Fitbit Charge 5. I plan on wearing it 24, 7 but mainly for my step count and heart rate. I would love input if anyone has used one.

    You bought a device without researching it? Or are you asking a technical question?
  • Ceriusly1
    Ceriusly1 Posts: 6,528 Member
    aundy28570 wrote: »
    I just purchased a Fitbit Charge 5. I plan on wearing it 24, 7 but mainly for my step count and heart rate. I would love input if anyone has used one.

    I have a Charge 4 which I suppose is different from the Charge 5 but I love my Charge 4 and I think it is very accurate on the step count and heart rate.
  • cwolfman13
    cwolfman13 Posts: 41,865 Member
    I have a Garmin Instinct. It is pretty accurate for steps as well as HR most of the time. Every once in awhile I'll get a weird reading for HR, but it's not a common occurrence. When I cycle I also wear a chest strap HRM that talks to my bike computer and it matches my watch the vast majority of the time.

    As for calories, these devices use a variety of inputs, population statistics and algorithms to estimate your daily energy expenditure...these devices are not sophisticated enough to provide you anything more than what amounts to a reasonable estimate. Things like elevated HR (for whatever reasons...stress, anxiety, overly hot, etc) can throw off estimates because your watch will think you're actually working harder than you are. As an example, a few weeks ago I had an interview for a promotion and was super nervous and anxious...for about 25 minutes before the interview my HR was around 125 BPM sitting in a chair in the waiting room. My watch recorded those as active minutes when in reality I was literally sitting in a chair going over my notes.

    In general my Garmin is accurate enough for energy expenditure for me to not completely dismiss it, but not accurate enough to say, "that's the number". My real world experience demonstrates that total daily energy expenditure from my watch is anywhere from 150-250 calories inflated depending on the day, stressors, weather and temperature, etc. It just gives me a number to work with and finagle per my actual real world results.
  • westrich20940
    westrich20940 Posts: 921 Member
    I think what you're really asking is: Is a heartrate monitor (particularly a wrist one) accurate for calorie burn.

    The answer is...they aren't. But there's not really a way to tell you how off they are.

    I used to use a Polar HRM (with a watch connected to a cheststrap). On a run it would give me a calorie burn estimate based on my heart rate during that run. My running app also would give me an estimated calorie burn based on distance/pace/my own stats/elevation change. One longer/more strenuous runs, the estimated calorie burn between my running app and my HRM could be up to 150 calories different with my HRM being higher.

    So, when I logged my calorie burn...I'd choose a calorie burn in between. For less intense workouts, it appeared to be more accurate.
  • ccrdragon
    ccrdragon Posts: 3,374 Member
    I have noticed that my Fitbit (wrist device) will grossly skew my steps count if I am doing heavy yard work like tamping gravel in a pathway or tilling ground in the yard. Funny thing I have also noticed is that it will count steps when I am riding the outdoor bike, but does not count any steps when I am riding the stationary bike.

    The heartrate function seems to be fairly accurate regardless of the activity - but I am fair skinned with not a large amount of hair on my wrists. I also like the watch to be snug because it is a pet peeve of mine to have it sliding up and down my arm.

Categories