Apple Watch Proves Most Accurate at Measuring Heart Rate in New Fitness Tracker Study

[pacalis]

So I read the article. Apple does something better than the competition. Then, just for laughs, I come here to the forum to read the howling and gnashing of teeth from the haters who want to tear them down.
I'm never disappointed, you guys are so entertaining! (& so full of it.)

The gnashing is that we don't know if they did something better than the competition. The presumptive gold standard, the Fenix 3 HR or newer versions aren't tested.

And we already know it's better than a basis peak because those are all recalled .

 

[notabadname]

Really, no Garmin Felix 3 or 5 with built-in HR monitor, waterproof to 300 ft, Sapphire available, 7 day battery.

 

[scaramoosh]

I've found the heart rate sensor to be anything but accurate.

 

[bobob]

My accuracy drops significantly when running, but when walking, or cycling, its pretty accurate.

How do you determine your accuracy?

 

[christian_koehler]

All optical sensors worn in the wrist have this problem sometimes.

With every step you run the device will move a little bit. When you run quickly at some point this 'signal' gets stronger than the oscillation caused by heart beat.

Or simply : The number you will see is your stride frequency instead of your heart beat. Some running watches measure cadence and heart rate and suddenly both are the same number. Usually something like 170 or 180 which is a typical value for steps per minute when running but unrealistic as a heart rate over longer periods of time for most people.

For me, compared to a medical stress test, I know that my 1st get Apple Watch consistently indicates about 30% higher when running. A knowledgeable friend said at my age, this the AW indicated rate were true, my heart would "explode."

 

[kdarling]

Apple puts a lot into R&D and calibration. They know they're the biggest target when it comes to accuracy. (...)
Here's an interesting read on the level technicality they went through to make it incredibly accurate at telling time. It's 4x more accurate than the iPhone actually. http://mashable.com/2015/12/30/apple-watch-synchronized/#EIku7j18gGq3

Which tells you how bad the iPhone's internal clock is.

You must not have been around for the discussions back when that PR smoke came out. It made it sound like the Watch was in sync with Stratum One time servers, when in fact Apple actually states that it's only kept with 1/20 second of the real time, which is several ten thousands times less accurate than those source servers.

Even to do that, the Watch requires constant updating via a connected iPhone, which doesn't say anything special about its own standalone timekeeping.

Now back to this thread, which is about wrist pulse reading (not actual heart rates, which are read at the heart itself).

 

[vmistery]

At the moment these devices are accurate enough for their intended purpose (i.e. non medial use). My apple Watch gives me an idea of what my heart rate is doing more accurately than I can count myself and gives me an idea if I can eat that burger at the end of the day :-D (Sadly the answer is all too often no).

 

[djcerla]

Because it would lose miserably!

Yeah, because Garmin and Sunto surely have more resources than Apple. And pink, magic unicorns.

 

[macUser2007]

Why don't these studies ever compare the Apple Watch with more hardcore fitness watches like those from Garmin or Sunto?

Yep, plus some of the devices in this study are rather old (the Gear S2 has been replaced by the Gear S3, for instances).

Here is a comparison with some other devices, compared to an EKG:

http://www.tomsguide.com/us/heart-rate-monitor,review-2885.html

RESTING:

BRISK WALKING:

RUNNING:

 

[iPadified]

I'm not sure why the study didn't include a Garmin product. This was how the trackers were selected:

"Following a comprehensive literature and online search, 45 manufacturers of wrist-worn devices were identified. Criteria for inclusion included: wrist-worn watch or band; continuous measurement of HR; stated battery life >24 h; commercially available direct to consumer at the time of the study; one device per manufacturer. Eight devices met the criteria; Apple Watch; Basis Peak; ePulse2; Fitbit Surge; Microsoft Band; MIO Alpha 2; PulseOn; and Samsung Gear S2."

It is an scientific study and not a review. I believe the scientists wanted to know how good these devices are. That it is not a comprehensive study does not matter scientifically. The method/paper also sets a standard for devices designers. That AW won has no significance scientifically.

 

[MentalFloss]

"Laboratory conditions". Great. My Apple Watch stops measuring my heartrate accurately when I start sweating. I think that condition would be more relevant for this kind of test than laboratory conditions.

 

[JaySoul]

In fairness, I have a Samsung Gear S3 and a monkey banging on a drum would be more accurate than the heart-rate sensor.

I even shaved my wrist to little avail. (Oh, the humiliation).

 

[t0mat0]

Anyone able to comment if the energy expenditure results show the Apple Watch as
consistently under/overestimating EE, or is the error under and over estimating?

I'm probably in favour more of accurate to slight underestimation if anything but curious. I think figure 2/3B has the plot.

 

[JustMartin]

Yeah, because Garmin and Sunto surely have more resources than Apple. And pink, magic unicorns.

It's not necessarily more resources, it's more experience in this area. Garmin and Suunto have been making dedicated running watches for years and they have been the gold standard for serious athletes for a long time. They have a reputation to maintain that goes above and beyond daily step and calorie counting.

 

[macUser2007]

Here is another test of various trackers against an EKG:

https://www.choice.com.au/health-an...ackers-with-heart-rate-monitors-what-we-found

"Of the 23 trackers equipped with a heart rate monitor, only nine had an acceptable margin of error when measuring active heart rate. Among the best performing were the monitors in the trackers from Fitbit, the Mio Alpha 2 and the Garmin Forerunner 235. These trackers managed to produce readings within 95% accuracy – despite having to contend with motion and sweat. The Samsung Gear S3 Frontier smartwatch was the best performing with an overall accuracy of 100%, but it can only read and display data when your arm stops moving. However, TomTom's Spark Cardio + Music and the Mio Alpha 2 return 99% accuracy even with movement, which makes them a much more effective training option.

Resting heart rate results fared better. We assessed these against industry-grade GE Healthcare Carescape V100, Dinamap V100 and Dinamap V150 heart rate monitors since first testing fitness bands in 2014. All models are calibrated and set to record data under the same conditions. The only difference is in input and export tools (the V150 is the first model to introduce a touch screen for example). 18 trackers fell within the acceptable margin of error, with the TomTom Spark Cardio + Music, Fitbit Alta HR and Garmin Vivofit Jr. all returning an overall accuracy of 99%.

Big electronic brands surprised. The Apple Watch Sport was accurate only 67% of the time, the Samsung Gear S2 53% of the time, and the Sony SmartBand 2 just 27% of the time."

 

[iCloudStrife]

Using the Apple Watch, I’ve progressed from running 8-10 kilometres per day to running 18-20. I still don’t care what my heart rate is though really most of the time. I assume it’s enough to keep me alive, which is really all I care about in the end.

 

[beanbaguk]

As many have already stated, it's very odd the likes of Garmin, Suunto, TomTom and Polar have not been included. These are the tools of most professional athletes.

 

[honglong1976]

Exactly. Garmin is such a giant in the fitness field, that to leave them out seems suspicious.

because Apple wouldn't come out on top and this was probably paid by them anyway! hehe Great marketing!

Also how old the fitbit surge?

 

[recoil80]

How do you determine your accuracy?

I press a finger to my neck and feel the heartbeat. Then I count the beats for 10 seconds and multiply the number by 6 to calculate the beats per minute.
It is not very accurate of course, but if the Watch reads 70 bpm and my calculation is above 100 bpm there is obviously something wrong.
It doesn't happen all the time, but sometimes the Watch doesn't read the right value, it is pretty obvious. If I'm running or walking pretty fast my heart rate cannot be low, but sometimes is 70bpm or even less than that.

 

[honglong1976]

I workout with two bands: Microsoft Band 2 and Fitbit Charge 2. When rested both shows the same HB but when Im working out its a totally different story. While my HB on MS Band 2 is 156, on the Fitbit Charge 2 is 135. In the end of one hour cardio workout I burn 800 calories with the MS Band 2 and 450/500 calories on the Fitbit Charge 2. I will only buy an Apple Watch when it has a continuous pulsing LED like the Fitbit Charge 2.

it is continuous during workouts! But that's it. That's one of the reasons I still use a Fitbit Blaze.

 

[Mausse]

Why don't these studies ever compare the Apple Watch with more hardcore fitness watches like those from Garmin or Sunto?

this! I just sold my apple watch series 2 (nike) it is unusable for serious running, intervales etc..

GARMIN is in my opinion still king

 

[44267547]

this! I just sold my apple watch series 2 (nike) it is unusable for serious running, intervales etc..

GARMIN is in my opinion still king

I do think Garmin has the edge over Apple in certain areas. . Especially for marathons. There was a fairly large marathon recently in my city and I saw very few Apple watches, mostly were Garmins.

 

[iBug2]

It's not necessarily more resources, it's more experience in this area. Garmin and Suunto have been making dedicated running watches for years and they have been the gold standard for serious athletes for a long time. They have a reputation to maintain that goes above and beyond daily step and calorie counting.

Heart rate tracking is fairly new. Garmin's expertise on running watches is not measuring your heartrate optically. So there's no reason why a Garmin watch should measure your heartrate better than an AW. It should track your cadence better than an AW yes, since that's what they have been doing for years.

 

[joueboy]

You want accuracy buy the chest band. They're not that expensive.

 

[Tech198]

Yay... "most accurate"