11.25: (Fake) Daytime nap replaces good night sleep data and tanks scores for previous couple of days. No way to correct after the fact.

Dr Phil said:

it could well be a factor and should perhaps not be so airily dismissed. 

It could but it isn't, as demonstrated by my last night.

Etupes25 said:

It could but it isn't

For you may be but that does not rule it out as a factor for others, which is probably one reason why Garmin include wearing the device properly for accurate detection of sleep. 
There are undoubtedly more factors involved than the position of the watch but the key to prompt resolution of any issue is ruling out elements that might adversely affect the end result. Knowing that you at least cannot wear the watch as Garmin suggest might be useful in resolving the problems you are encountering with sleep determination. And if there are others who cannot wear the watch properly and are also having problems this might give Garmin some direction. Not helped much of course if people who are wearing the watch properly also have problems. 

I think I'm wearing my watch correctly. For me, the false sleep detection has also been present on the Fenix 6 from time to time (or FW version to FW version), but--as mentioned--you could usually spot the problem, because the watch would display "Sleeping". Secondly, you could also edit and sync sleep times, which does now only work if it wants to (which isn't often and I don't know why it sometimes works). That's my main gripe (I know, I sound like a broken record). If I had the option to manually correct the mistakes made by the watch, I'd just shrug it off. For the moment, I can't wear the watch continuously or have to deactivate the OHR sensor on the couch to avoid messing up Training Readiness etc. 

[Redacted] said:

I can't wear the watch continuously (or have to deactivate the OHR sensor on the couch) to avoid messing up Training Readiness etc. 

I am here as well. I am going to try a different device for sleep analysis and see if it is worth it. I need good sleep analysis to evaluate whether the various settings of my CPAP machine are impacting positively my sleep scores. I run a one-week trial with a certain settings configuration, collect the sleep apnea data, the sleep cycles data and the sleep score, and put this in a nice excel model. THen I change one of the many settings. I have been building my data model over the last few months and lost quite a bit of data accuracy since my my sleep data gets screwed up at least one or 2 days each week.

The sleep detection is really a rogue algorithm on the watch and/or doesn't provide the hooks for the UI to enable user control.

This is a problem that continues. Please vote for resolution in the beta section - Faulty short sleeps replaces long correct sleep - Public Beta 11.xx - Fenix 7 Family - Public Beta Reports - Garmin Forums

Well, in fact Garmin somewhat knows you are taking a nap, because it will show in Body Battery as long as the nap is long enough to be registered. In some instances you can actually see your BB to go back up or at least plateau for the duration of the nap. And that, togetherwith a most likely lower HR and lower stress, should be beneficial for your Recovery time.

I seriously wouldn't recommend what you are trying to do. your CPAP machine should provide enough data to confirm if its set up is correct.

Watches cannot correctly determine sleep stages. Only thing from your watch I would consider checking against the CPAP is your HRV and maybe restless and awake to see if that correlates.

Garmin Deep sleep is OK but its light and REM is rubbish - which is entirely understandable as a EEG device and looking for specific body reactions is required to appropriately identify these stages.

EEG devices are only up to 80% accurate, the withings sleep mats are about 75% accurate (based on the last hospital study that was done on them). Watches done on dubious testing (due to data collection being far too small) only puts best stage around 80% to a max 80% accurate device. i.e. do not rely on any watches sleep stage data at this point in time!

firemaker69 said:

Everyone knows that a nap during the day is great for you........except Garmin.

Depends on what you consider a nap. There is no defined benefit for a nap greater than 20 mins (based on imperical studies). Naps aren't supposed to be over 40mins - so actually Garmin incorporating naps into body battery makes far more snese, as it would skew sleep score.

So no these short sleep sessiosn should be ignored other than improving body battery.

jaamgans said:

I seriously wouldn't recommend what you are trying to do. your CPAP machine should provide enough data to confirm if its set up is correct.

Don't worry, I have this down. I am using the OSCAR software to track all the sleep apnea settings and events. The only thing that this doesn't include is a sleep quality factor. There are various settings (like response type and expiration pressure relief) that have a mild influence on the eradication of sleep apnea events with some compromise to be made, but they influence the comfort of the therapy. So I am tracking 3 settings in combination and their effect on 10 CPAP data points, to which I add the sleep score as extra for sleep quality.

jaamgans said:

Only thing from your watch I would consider checking against the CPAP is your HRV and maybe restless and awake to see if that correlates.

No, I don't need to evaluate the results of the therapy through the watch, I just need the watch to give me a quantified data point about sleep quality. I could go with using a feel-based rating (like athletes complement training effect with RPE).

jaamgans said:

do not rely on any watches sleep stage data at this point in time!

Mostly this watch doesn't count awake time correctly, but replaces it with REM & light. But it doesn't matter if it does it consistently: I am looking for how the sleep score evolves in average from week to week, a bit like the HRV measurement does.

I just want the watch to deliver what it is supposed to deliver: a reliable, if not perfect, sleep score that corresponds to my nights, not to non-existent naps.