[POLL] OHR plausibility when not exercising

talsela said:

Furthermore, most people who complained about their OHR usually showed some kind of test or some kind of comparison, whereas I haven't seen a single post about OHR working great (not during exercise) and base it on anything other than their sheer feeling

People that post videos with problems do so to show what they are experiencing in the hope others will understand what's happening and could possibly assist with a solution. Why would someone take the time to create a video showing there's nothing wrong with their watch?  They would have no need to prove to you that their watch works fine.  How would doing so benefit anyone?  Are you insinuating by not doing so and asking how confident they are by their poll answers that they are oblivious to a defect or worse, lying? Do the videos fronius3 posted satisfy you?  If I take the time to conduct your test and post a video for you to see, how does that fix your problem? 

I'm not talking about a video. Showing something works for a few seconds doesn't mean it's fault-free. Here's my OHR behaving exceptionally well:

At other times it doesn't, and I don't always see it nor do I always recognize it's wrong. As I've already said - looking at it and seeing 66 I can't tell for sure if it's correct or as much as 30% off.

Yes, I do think there may be some who believe their OHR is perfect only judging by random glances at the watch and still be oblivious to inaccuracies because they (a) don't check all the time - nobody does (b) can't always tell when a value they see is innaccurate.

Ok, so I conducted a test this evening as best as I could to compare the OHR in my Fenix 6 Sapphire to an external strap when the Fenix is just in regular watch mode.  Since I do not work for Garmin and have no access to design documentation, I had to make some assumptions here:

1) The fit file in the MONITOR directory (hereafter referred to as the "monitor fit file") is the HR data stored by the watch during regular mode for the last few hours.

2) The data in the monitor fit file is stored in GMT time (I am pretty confident about this based on my tests)

I put on a Garmin HRM3 heart rate strap with conductive jelly since I was not doing any sport and did not want to worry about poor contact due to the strap drying out.  I paired it to my Edge 520 Plus and ran a 30 minute activity with the Indoor Cycling profile.  I also specifically disabled the HRM3 in my Fenix 6 even though I didn't expect it to use it just to be 100% sure.

During the 30 minute time period I was recording on the Edge, I did everything normal sitting outside, playing fetch with my dog, walking around the back yard picking up the dog's mess, etc.  I also ran back and forth in my back yard with my dog to get my heart rate up for a short period.  During the entire time, I did not touch any buttons on my Fenix 6 - it was just displaying the watch face the entire time.  Also, I do not have the HR metric on my watch face.

The graph from the comparison is below:

Green dots are from the monitor fit file from the Fenix, blue line is from the HRM in an activity on the Edge.  You can see many of them are close and even right on target.  Others are quite a bit away, and it completely missed several spikes.

While I am a bit disappointed, I actually am not that bothered because I will start an activity for pretty much anything other than going around the block for a walk with my wife (and I've been known to start activities for that as well).  I can post another graph is anyone cares that compares my Fenix 6 OHR to an HRM strap when I am in an indoor cycling activity on my Fenix 6 and the OHR tracks pretty much perfectly, but this thread is specifically about outside of activity so I will not post it.  Also, for nearly all activities where I care about HR I will wear a strap because I have done so for the past 10+ years and it is not uncomfortable to me.

I am happy to send anyone the raw data if you want to see for yourself.  I used the Fit File Repair Tool to extract the data from both the monitor fit file and the activity fit file and merged them in Excel for anyone who wants to try the same thing.

Whenever I upgrade my Garmin, I sell the previous one so I do not have any others with the older OHR chipset to compare to - sorry.

Also just to be clear, I am probably the poster child for an OHR sensor to work perfectly - I am light skinned, have very thin and light arm hair, no tattoos, etc.

who knows said:

You post a poll and when that poll doesn't go in your favor

First of all, I believe we can both agree that the sample size is still too small. But since you commented on the direction of the poll anyway, I have to ask you what makes you think its direction is a good one? ~40% claim they see OHR values that don't make sense to them either quite often, most of the time, or all the time.

talsela said:

I have to ask you what makes you think its direction is a good one? ~40% claim they see OHR values that don't make sense to them either quite often, most of the time, or all the time.

As I said, I'm also looking at the same poll you have in the 945 forum.  Same sensor and you've left those results out.  Maybe add this poll in the 245 forum?  And I say it's not going in the direction you want not based on percentages, but because you're only questioning the answers given by one group and not the other.  You accept the 40% as correct replies, but the other 60% are questioned on their validity.

The reason I left the FR945 poll results out were because of the sample size. When I started the polls I thought there would be a couple of hundred of answers, not a few dozens. I also don't think you can just add these results up because there are bound to be some who voted in both. But I'll humor you and do it anyway, adding the 50 results here to the 14 there:

1. Never - 21.9%
2. Hardly ever - 28.1%
3. Somtimes - 15.6%
4. Often - 12.5%
5. Most of the time - 15.6%
6. All the time - 6.3%

So only half of responders are in the "never" or "hardly ever" category, and over 1 in 3 claim to see "wrong" values either quite often, most of the time, or all the time. You decide if that's good or not.

The reason I asked only the "accurate" group is because I assume most of us trust these devices by default, and to say yours shows funky values probably came after getting a good reason to check, like seeing 90bpm while sitting on the couch or no higher than 70bpm after climbing 5 floors.

For me personally, once I started checking, I noticed that even the plausible values are not necessarily always correct. Even 63bpm while sitting at the computer - which I would have never doubted before - is 20% off from the actual 51bpm.

who knows here's an example from today of someone who claims their OHR is not accurate. I definitely don't need to question him because he's already questioned himself and tested and proved his hypothesis.

who knows said:

Maybe add this poll in the 245 forum?

Well, there are at least two threads about issues with the 245 OHR at rest:

1. Abnormal wrist heart rate peaks while resting
2. Wrist Heart Rate Monitor Not Working at Rest

Hard to say, but doesn't seem unreasonable. When I connect my chest HR it seems to be more or less the same HR.