Watch on the charger but it thinks I'm in deep sleep. Bug?

Garmin is aware and I don't think they will fix. The sleep processing is done through the accelerometer and then post processed on their server. If the watch doesn't move in your sleep time it assumes you are asleep. Either live with it (this is what I do), or charge during the day ie not in the specified sleep time. Those are your options.

HTH

CW

Garmin Connect shows that I was in "Deep Sleep" during the charge period. That's a pretty serious bug. If the watch is being charged, it CANNOT be on your wrist.

Not being on your wrist doesn't mean you weren't asleep, either. The 24×7 activity tracker aspect of the watch does not (and was not designed to) only report energy consumption – including due the basal metabolic rate – and sleep that it can ‘prove’ to have occurred through conclusive (data pertaining to) physical evidence obtained by its own sensors, and there is a large component of guesswork involved. Sometimes it just guesses wrong.

When I started using my FR235, I also felt it was wrong and ‘stupid’ that Garmin Connect would report charging time at night at sleep time – which it sometimes does – but then I realised:

• Without actively monitoring the user's brainwaves, the sleep (or 24×7) tracking device will always be guessing – with various levels of confidence around the clock – through circumstantial evidence that indirectly informs it of whether the user is asleep.
  
• In the absence of a state of I-don't-know (or I-can't-tell), alongside Asleep – with or without gradations on quality of sleep – and Awake, in the way sleep tracking/reporting is implemented, logically it is inevitable that there will be errors, especially when the tracking device is not in a position to gather data.
  
• We, as users, are not collectively or individually asked to state our preferences for what to do when the watch does not know if we are sleeping or not, so there is no ‘bug’ that can be evidenced by Garmin Connect either ignoring or overriding our stated preferences.
  
• Not giving the user a say is by design, and not a ‘bug’ by any stretch of the imagination, no matter how you and I may feel about it. Respecting and doing what the user prefers, even when the tracking application is knowingly imperfect, is not a requirement.
  

Anyone else have this issue?

Yes, from day one.

However, I have also (since) seen occasions when charging during my normal sleep hours not being reported as sleep. On some occasions, Garmin Connect even revised its assessment; when I was charging the watch from 1AM to 3AM, Garmin Connect reported that I was sleeping at the time, but when I disconnected the charger, and then put on the watch and went to bed from 4AM to 10AM, in the morning the sleep chart was revised, and the period between 1AM and 3AM was no longer shown as sleep.

Is Garmin aware of it?

I'm confident that is the case.

I have a feeling I got a dud.

I won't comment on the battery life aspect, but if you feel that a 24×7 tracking device is not fit for purpose if it attributes sleep on occasion when it's not being worn by the user, then return it as not fit for purpose, as opposed to try to change the device (or the manufacturer's design principles) to suit your personal preferences as if they/you matter. Your (and my) prerogative and ‘power’ as a consumer is in choosing among imperfect compromises on offer to us in the market, not securing perfection and/or satisfaction.

Either live with it (this is what I do), or charge during the day ie not in the specified sleep time. Those are your options.

Or give up but choose not to live with the FR235's and Garmin Connect's behaviour in that regard, and buy something else and live with an alternative compromise instead.

Thanks for responses

Hey guys,

Thanks for your info. I called support today and they said that indeed the determination of sleep is done on the server done with accelerometer data.

So it's not a bug, but it's a serious shortcoming. There's a heart rate monitor (also sending data to the server) that could be cross referenced.

If no movement and no pusle, owner is dead or unit is off-wrist. You know what they say about a resting heart rate of zero.

Regarding the battery, they say it's not normal and will exchange.

Cheers,

Murk

The sleep tracking is really my only gripe at this point.

It should be smart enough to check that there is no heart rate or movement being recorded and know that the watch is not being worn. If there is no movement and no heart rate, don't assume the user is asleep.

If there is no movement and no heart rate, don't assume the user is asleep.

That would be my personal preference, but Garmin Connect equally ought not assume the user is awake either, just because he/she is not wearing the watch, especially when that happens during user-specified normal sleep hours. In the absence on a third state in a sleeping-awake dichotomy, GC has to make a call one way or the other – and, in that regard, a user must not assume Garmin is obliged to take into account his/her personal preferences as if they inherently matter.

Yes, I'm arguing for the logical position of Garmin not having to be bound by my personal preference as a user and its customer. Just because I'd like something, it doesn't automatically mean I think others have to accommodate me.

That would be my personal preference, but Garmin Connect equally ought not assume the user is awake either, just because he/she is not wearing the watch, especially when that happens during user-specified normal sleep hours. In the absence on a third state in a sleeping-awake dichotomy, GC has to make a call one way or the other – and, in that regard, a user must not assume Garmin is obliged to take into account his/her personal preferences as if they inherently matter.

Yes, I'm arguing for the logical position of Garmin not having to be bound by my personal preference as a user and its customer. Just because I'd like something, it doesn't automatically mean I think others have to accommodate me.

Why does GC "have" to make a call one way or the other? I agree that makes sense when the watch is detecting little or no movement. My wife is one who appears to be in a coma while asleep. I disagree when there is no HR detected. There is a third option Garmin can use and I believe most user's would agree on. Just like the HR graph displays no data when the watch has been removed or turned off, the same should be displayed for sleep regardless if it's during your programmed sleep window or not. No HR, no data.