Scientific Review - Many Garmin Watches Extremely Inaccurate

This is an intense conversation 

I question whether it’s normal to buy a 1000 USD device and not care if the features work as advertised. If the dealership sells me CarPlay, I expect to be able to connect my iPhone to my car. 

I’m not sure what you mean by paranoid. The dude is a scientist who’s into biological informatics. He writes code to analyze this type of data, so it makes sense that he might be into testing the accuracy of devices that measure biological functions. There’s nothing paranoid about calling out companies for selling subpar products. It’s part of what motivates them to do better, especially with Garmin when Apple is not far behind. I like the Garmin ecosystem better and I want their devices to be accurate and succeed.

He uses known to be accurate devices such as brain MRIs, EEGs, etc to compare the devices against to provide unbiased accuracy results. Could he do better science? Yes. For example, he could test on more persons than just him.

What he’s doing makes sense to me. What doesn’t make sense is you rattling off at the mouth with nothing to back up anything you say. Again, where is your data analysis to suggest he’s wrong? Post your data analysis here in the comments. I’m waiting.

The Garmin EPIX accuracy is highly questionable. There is not much data that someone without a load of other equipment can verify, but respiration rate and when you are asleep or awake are two bits of data that can be verified.

For respiration rate you can do a health snapshot and control your breathing rate to a consistent level for the duration. I have tried at 6brpm and at 30brpm. The device records about 14brpm for both… totally inaccurate and useless.

For sleep tracking it sometimes misses periods when I have actually woken up. Last night for instance I woke at 4am and got out of bed to get a glass of water before going back to bed and sleep again. The sleep tracking missed that completely. Now if it can miss me climbing out of bed and walking to the kitchen I cannot believe it can be recording the more subtle nuances of sleep correctly,

So the two features I CAN verify either don’t work at all or appear to be woefully inaccurate and inconsistent. That gives absolutely no confidence in the accuracy of the other features which I am unable to test with out a houseful of control equipment.

I would recommend that people who wish to criticise a YouTuber for daring to suggest Garmin watches are inaccurate tries the health snapshot respiration test twice at a controlled 6brpm and 30brpm. Still feeling confident in your device accuracy?

Garmin need to sell watches that actually have working features rather than constantly adding more features which don’t work.

Respiration rate algorithm doesn’t work as you think. On my runs works fine. Sleep is accurate in my case. Yea Garmin must make a lot of improvements but again is a sport watch.

Respiration rate doesn’t work… regardless of what I think

It shouldn’t matter if it is a sports watch or not (although this forum appears to be for outdoor recreation wearables, not sports). If it is sold with advertised features they should work correctly.

Respiration rate works fine on my runs 

Two things. 

1 - How do you know it works fine? Have you counted breaths during the activity and are sure it is giving you your correct respiration rate and not just a pre-set approximation based upon your level of activity?

2 - You will be using a chest strap and not the watch to gather the data as the watch cannot record respiration rate data by itself during an activity.

I’ve already emailed Garmin support multiple times and have an active thread about this in these forums. I believe it has also been posted in the developer forums. Last response I had from Garmin Support was 10 days ago which indicated they couldn’t answer whether this was a problem they could replicate (why, I have no idea. It would only take 4 minutes to verify for them!)

I suspect it may become yet another unanswered problem which is given no attention whilst they concentrate on introducing a new cake making or knitting activity to the Epix Gen 3.

Can you share a ticket number so I can ask them to add me to the tracking case? I am actually honestly surprised just me and you noticed this. I actually originally noticed this on a run one day where the watch said I was at 35 brpm, and I definitely wasn't breathing that fast as I have pretty controlled breathing. The only time I could get a "different" breathing activity is when I did the breathwork activity and defined in the watch the 4-4-4-4 method. Really odd and kind of frustrating. I think another forum member commented that the brpm was "encoded into our HRV" and that's how the watch picks it up but I think you're right that it guesses based on heart rate or something.