Autotune Insulin and Carb Sensitivity at different times of day.

[ChiSoxDev]

Is your feature request related to a problem? Please describe.
For nearly everyone insulin sensitivity changes throughout the day, but autotune only provides one value. Documentation says "For ISF, it calculates the 50th percentile (median) deviation for the entire day and determines how much ISF would need to change to get that deviation to 0. It applies 10% of that as an adjustment to ISF."

Describe the solution you'd like
A good first step would be splitting the data against three or four time intervals throughout the day to provide three different numbers. I propose 4am - 10am (morning), 10am - 4pm (mid-day), 4pm - 10pm (evening), 10pm - 4am (night).

Describe alternatives you've considered
Individual experimentation will always remain an option, but this is the type of use case autotune exists for.

Additional context
I appreciate the consideration.

[ChiSoxDev]

Edited to have this include Carb Ratio (Sensitivity) as well.

[scottleibrand]

We looked into this a few years back and determined it wouldn't be practical to try to calculate different ISF values for different times of day: there simply isn't enough non-excluded data for many time periods to get a reliable signal out of the noise.

It might be possible to do for carb ratio, as that is a more direct calculation of BG rise for each meal. But someone interested in developing and testing this would need to coordinate doing so, and so far we have never found anyone that interested in building a solution.

[ChiSoxDev]

That makes sense. Would you be able to direct me to the Autotune code that specifies the logic used? I am not a developer, but with my experience as a Windows Engineer I'd like to extrapolate on it further using Power BI to see if I could provide the data insights you weren't able to previously find.

Also, can we leave this request open for someone to pull at a later date to work on the carb ratio?