Why are classic, summary, awake minutes not equivalent to minutesAwake?

[sleepy_dave]

I'm exploring summary data within the levels field of classic sleep, and have found that the values for awake minutes is not the same as those in minutesAwake (screenshot below). Oddly, the same summary values for wake minutes in stages sleep is equivalent to minutesAwake for those records. Why does classic's awake not match up? Is there a difference between classic's awake and stages' wake features?

[GordonFitbit]

Hi @sleepy_dave.   We have several endpoints for fetching sleep data.  Would you tell me which endpoint you're executing?  

[sleepy_dave]

Hi @GordonFitbit, thanks for the reply! I'm using the Get Sleep Logs by Date Range endpoint described here.

[GordonFitbit]

Hi @sleepy_dave.   

A "classic" sleep means that its v1 of our sleep and "stages" is basically sleep v2 (it comes from our relatively new trackers). With "classic" sleep, we detect a person is asleep through movement.  If the person hasn't moved for about an hour, the device will assume you're asleep.  Additional data, such as length of time of your movements, is an indicator of sleep behavior and help confirm when you're asleep.  However, if you're not moving but not asleep for long periods of time, it's possible for the tracker could falsely record sleep.

When recording sleep "stages", Fitbit uses a combination of movement and heart rate patterns to estimate sleep stages.  Therefore giving a more accurate recording of sleep / wake information. 

https://help.fitbit.com/articles/en_US/Help_article/1314/?q=classic+sleep&l=en_US&fs=Search&pn=1#how...

[sleepy_zank]

Thank you, @GordonFitbit. I've read through the docs you've provided regarding classic vs. stages sleep and am still a bit confused. In the screenshot above, I have a list of records from classic stages. There are two features that seem to describe time awake over the course of sleep. The first being "minutesAwake" and the second being "awake_minutes" within the "summary" feature. Can you help me understand the difference between these two features?

Edit: Sorry, I've been logged in on a friend's dashboard looking at his data so I've accidentally responded from his account.

[GordonFitbit]

Hi @sleepy_dave.  The minutesAwake value should be the sleep period (or "duration") minus the minutesAsleep value.  The "levels:data" section show the raw data for the 3 sleep stages (asleep, restless and awake).  The "levels:summary" summarizes the "levels:data" for the 3 sleep stages.  Therefore, the minutesAwake should be closely related to the summary's "awake" + "restless" values.

[sleepy_dave]

Hi, @GordonFitbit. I've added the "levels:summary:restless_minutes" values to a similar screenshot as I provided originally and your instructions match up for the most part, so thank you! There are two rows below where that is not true, however: 2018-08-19 and 2018-08-13. Can you help me understand why summary's "awake" + "restless" might end up being greater than "minutesAwake"?

[GordonFitbit]

Hi @sleepy_dave, 

Would you please private message me the user id for the Fitbit account that's collecting the data?  You can find it be clicking on the profile icon at the top right corner of the Fitbit dashboard.

Thank you,

Gordon

[sleepy_dave]

@GordonFitbit Just sent! Thank you.

[dp_dev]

Hi Gordon, I just have faced with the same problem (minutesAwake != summary awake minutes + summary restless minutes).

Could you please kindly explain why this is the case? We suspect, that one or two "edge" segments are not counted into minutesAwake (by "edge" segments I mean the very first and/or the very last awake/restless segment of sleep data). But I'm not sure and couldn't find any proof.

Thank you.

[GordonFitbit]

Hi @dp_dev 

Would you please private message me the entire sleep log that you're looking at?   I'll see if I can figure out why the data is not adding up.

Thanks!

Gordon