First evidence (from Apple) the HR sensor may not be accurate

[Julien]

It is common knowledge that optical HR seniors have poor accuracy with dynamic moments and escalating HR. We have all been hoping Apple has perfected the process a lot like they did with the Finger Print reader. However the latest post on Christy Tulington's blog may show otherwise.

She says she has done 2 runs since getting back. She said she did an 8 miler and a 16 miler, plus a bootcamp session. There is a pic of her Watch with the caption "A summery of my evening run". At first glance it sounds like her 8 mile run but the time is far too short for her speed. She is not a 7:30 minute mile runner over distance since she ran 9:19 minute miles in the ½ marathon. However even for her small frame the quantity of calories burned in an hour is far too low for a run at only 298kcal (plus 72kcal resting). If she just walked an hour she should burn this many or more. Maybe this is actually from her bootcamp session which would sound more plausible but still low.

Not sure what to make of all the discrepancies but the calorie count for time is low and may indicate inaccuracies in the HR sensor.

 

[jabingla2810]

Calories burnt is calculated a variety of different ways by a variety of different fitness systems.

I don't think this finding gives us much indication of the accuracy of the HR sensor.

 

[Ries]

Calories burnt is calculated a variety of different ways by a variety of different fitness systems.

I don't think this finding gives us much indication of the accuracy of the HR sensor.

A 8 mile run would burn at-least twice that, even with her low weight.

 

[Julien]

Calories burnt is calculated a variety of different ways by a variety of different fitness systems.

I don't think this finding gives us much indication of the accuracy of the HR sensor.

Well aware of that and have used many different systems over the years. FirstBeat has arguably the industry's best algorithms for calorie calculations. Not sure what Apple is using.

While circumstantial the amount burned for effort is still seems low. Also IF it is an 8 mile (7:30mm/8mph) run (which I doubt based on her posted abilities) it is far to low.

 

[jabingla2810]

A 8 mile run would burn at-least twice that, even with her low weight.

Absolutley true.

But where does it state the calories burnt was calculated using the HR sensor.

The treadmill I use at the gym doesn't have a heart rate sensor, but it has a go at calculating calories burnt.

I agree, these figures seem odd, but it doesn't necessarily point to the fault of the HR sensor.

 

[FatMax]

That screen sure looks gorgeous though!

 

[cmChimera]

I doubt it's an actual summary of any long run. It's much more likely that it's a shot just to show the interface.

But assuming it is true, she may not have the HR sensor on and the distance/step counter is ****.

 

[Julien]

Absolutley true.

But where does it state the calories burnt was calculated using the HR sensor.

The treadmill I use at the gym doesn't have a heart rate sensor, but it has a go at calculating calories burnt.

I agree, these figures seem odd, but it doesn't necessarily point to the fault of the HR sensor.

A down and dirty on calorie calculation. A treadmill uses fixed weight and speeds algorithms to (poorly) estimate calories burned.

A watch with HR monitor adds by far the most important and determining factor. With know weight/age/sex/MaxHR/MinHR then your HR is directly correlated to the number of calories you burn. So if it IS reading low it IS directly related to the HR monitor.

Granted we don't know the full extent of the info but it seems odd that Apple would post this with the discrepancy it implies.

----------

I doubt it's an actual summary of any long run. It's much more likely that it's a shot just to show the interface.

But assuming it is true, she may not have the HR sensor on and the distance/step counter is ****.

It is still an hour of activity.

Also are you saying you can cut the HR sensor off while recording a workout? I didn't see this in the Companion App and what is the point in having a HR monitor for exercise if you don't use it?

 

[cmChimera]

It is still an hour of activity.

Also are you saying you can cut the HR sensor off while recording a workout? I didn't see this in the Companion App and what is the point in having a HR monitor for exercise if you don't use it?

You can shut it off, and there is a disclaimer that calories burned, etc. would be less accurate. I can't pull up the picture now but I'll post it if someone worse doesn't later. I agree that I wouldn't turn it off but it's an option.

 

[JayLenochiniMac]

Apple is Burned.

 

[Mr. Buzzcut]

The watch really isn't bringing anything to the table for fitness use. If they were more serious about it, they would have traded the OLED screen for a GPS. Instead of focusing on data and results, they went for eye candy. Maybe that's what most people will want. But there seem to be quite a few, myself included, who were looking forward to the fitness potential and for Apple to do more than other wearables in that arena.

Maybe you'll at least be able to pair a chest strap for vigorous workouts and use the built in sensor for periodic checks the rest of the day.

 

[cmChimera]

Old screenshot but I doubt they removed this feature. Apparently it's in accesibility.

 

[Runt888]

If they were more serious about it, they would have traded the OLED screen for a GPS.

It would be really hard to use the watch without a screen

 

[Mr. Buzzcut]

It would be really hard to use the watch without a screen

I get you're kidding but the watch could have been just as useful with a lower power display. Maybe even more usable (sunlight).

 

[Defender2010]

It's quite obviously not an 8 mile run. (Even the OP doubts it is) This gives no indication that the HR sensor may not be accurate.

 

[Patriot24]

The photo directly adjacent to the shot of the watch has the caption "A few miles before putting my kids to bed." It is likely that the data on the screen is from that short run. Perhaps "9:35" is the time?

 

[Mr. Buzzcut]

The photo directly adjacent to the shot of the watch has the caption "A few miles before putting my kids to bed." It is likely that the data on the screen is from that short run. Perhaps "9:35" is the time?

Sure looks like 9:35 is the time of day and one hour for the activity time under "Summary."

----------

By the way, stated battery life of the Fenix:

Up to 50 hours in UltraTrac mode; up to 20 hours in GPS training mode; up to 6 weeks in watch mode

https://buy.garmin.com/en-US/US/on-...ith-metal-band/prod516264.html#gallery-dialog

(sorry, meant for this to fall under my previous post)

 

[Mascots]

Is that too low for a paced run?

I wouldn't think so for her activity level and weight. While it may be low, I find it within perfectly acceptable limits for the shape she is in.

Edit: misread. Hmmm.

 

[M-5]

Apple could have just as easily rendered that screen in photoshop to make it look pretty with different readings.

 

[Multiverse223]

Not that anything you say really means anything because of such a lack of information, but calories burned is not a very useful metric, and it's a very rough estimate. It's fun to get that estimate, but it's not going to do anything more than offer motivation to people. The complexity of human metabolism is far and beyond greater than calories in vs calories out. First of all, it varies greatly from individual to individual, and how fit that particular individual is, down to certain muscles. For example, someone just starting out will burn more than a well trained individual because their body becomes more efficient at it, and thus uses less energy to do the same task.

Moreover, calories burned doesn't tell you what macronutrients were oxidized for said energy expenditure. For the most part, unless the individual is fat adapted, they'll be oxidizing glucose, and not fat, unless they continue to exercise at a moderate intensity for over 20min. Anyway, studies have shown exercise doesn't do much for weight management.

Exercise is great for many things, but nutrition is far more important in general, and especially for weight management. Nutrition is probably about 80% of weight change. People need to be made aware that they shouldn't consume hardly any sugar (only from some fruit really, not fruit juice either), and no highly processed foods, and generally consume less carbohydrates, if they wish to lose weight or be more healthy.

 

[Lobwedgephil]

Thats just a short evening run, as she says, so see zero evidence of this showing the HR sensor isn't accurate.

 

[8CoreWhore]

http://help.fitbit.com/articles/en_US/Help_article/Heart-rate-FAQs

Interesting info - some of which may pertain to the AW. Scroll down to "WHAT IMPACTS THE ACCURACY OF MY HEART RATE READING?"

They say you have to wear it at least 1 fingers width below (towards elbow) your wrist bone. That it needs to be snug but not too tight, or it won't be accurate. And if even further towards the elbow, it will be more accurate still (so not totally accurate at 1 fingers width!).

Look at these pics of AW's how far from the wrist they are... The one OP posted... Looks kinda loose, actually... See how off-centered it is?

I have to say, the AW sensors are HUGE compared to the competition, so they may be the most accurate.

 

[saving107]

Apple could have just as easily rendered that screen in photoshop to make it look pretty with different readings.

Evening run at 9:35 (doesn't specify AM or PM). If its PM it looks really sunny for it being so late, if AM then why is it a summery of an evening run?

I agree with you, this is more likely just a photoshop image for promotional use.

 

[matrix07]

Nice investigative work. Looks like she's using SS with white band.

But I don't think that's the photo from her run. Looks too good for self capture.

 

[zacheryjensen]

I do not think this screen is what you think it is. I think it's a summary of the entire time the watch has been on since it was last removed and then worn again. Saying it's been on an hour, calories burned active vs. resting. If it's even remotely honest, it was likely a short run surrounded by mostly resting and the watch has only been on the wrist for an hour.

I'm far more likely to believe it's just a staged shot, or photoshop substitute.