This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Night Sleep Data gets replaced by daytime "sleep"

The problem is that in some cases a normal sleep recording is replaced by a very short nap and totally screws all connecting data like the recovery status.

Garmin now says that this is by design, which makes no sense at all.


  • How long was your nap? I am being told a nap of 3 hours or more should replace your prior night's sleep. I am further researching shorter naps still being picked up that are replacing your sleep.

  • I agree with you. The watch sleep detection is completely out of control of the user and it does not do a good job. The user should have a way to start and stop automatic sleep detection, to trigger nap or sleep detection when intended, and to avoid all these false positives that ruin the sleep scores. Sleep should be additional, and therefore naps should not replace the previous sleep period, but add to it.

    The situation is idiotic for me. I have to be careful about removing the watch to avoid false sleep detection during the day.

  • My nap was approximately 2 hrs 15 min long. It was replaced with a zero sleep score. And when I manually changed my sleep time to match my night sleep, my score was 49/100. My original sleep score was 77/100. So even changing my sleep hours manually is lowering my sleep score by a lot and I cannot find any solution or workaround for this except removing my watch if I suspect I will fall asleep outside my regular sleep hours. By doing this, I will miss on the increase in body battery during sleep. This situation is far from ideal. I believe you that this might be by design, but i find it difficult to make sense of the logic behind this implementation.

  • Thank you for the additional information provided. I am gathering more information from all new reports that are being made within this thread. If you have a nap less than 3 hours that has replaced your prior night's sleep, please highlight my name and send me a Private Message. Thank you!


    For those in this thread that have been personally attacking and had your posts deleted, I will be happy to either put you on moderation and we will need to manually approve every post you make or I will outright ban you.

    Please consider how few companies provide worldwide public forums. If you find yourself bored and your only contribution to our forums is to generate a little chaos for amusement, please reconsider posting. Thank you.

  • What was your heart-rate during that time, and what is it typicality during a real night of sleep?

  • The heart rate at the time was the same as it was in the morning when I was working. At night she was calmer. The watch should never have recorded that as sleep. Above all: The clock itself says I was awake at 2:35!

  • Interesting. For me it works really well, sitting still at the computer never triggered a sleep event for me. Even small 20min naps never did. A few days ago I had a sick day, got a decent sleep recording for the night but later went back to bed from late morning till afternoon. That was the only time in my Garmin history I got a day sleep recorded, it removed my night sleep, I manually put in my night sleep time again.

    I really wonder what triggers this. But it seems on the timeline you have a big drop in your heart-rate in the beginning that the watch could have interpreted as sleep. Then with the larger spikes it logged you as awake, then sleep again. Its a lower heart-rate than you had in your morning it seems. I kind of can understand why the algorithm could think that was your main sleep in the afternoon.

  • Interesting. For me it works really well, sitting still at the computer never triggered a sleep event for me. Even small 20min naps never did. A few days ago I had a sick day, got a decent sleep recording for the night but later went back to bed from late morning till afternoon. That was the only time in my Garmin history I got a day sleep recorded, it removed my night sleep, I manually put in my night sleep time again.

    I really wonder what triggers this. But it seems on the timeline you have a big drop in your heart-rate in the beginning that the watch could have interpreted as sleep. Then with the larger spikes it logged you as awake, then sleep again. Its a lower heart-rate than you had in your morning it seems. I kind of can understand why the algorithm could think that was your main sleep in the afternoon.

    That's a typical day for me, lowest heart-rate during the night, sleep was detected correctly.

    This was my "sick day" little higher heart rate during the night cause I slept badly, but got registered as sleep in the morning. Later my real day-sleep, way lower hear-rate than during the night. Software later decides to make that my main sleep until I edited it back.

    So yes a daytime sleep should not delete the real night sleep, you should be able to get that back. But from the algorithm standpoint this would be the interpretation of a night worker sleeping in the day, purely based on heart rate.