Where does navigation course finish time estimate comes from?

You mean in the watch while running?  I think it's a combination of the time data in the file, current pace and distance to destination. 

You mean the ETA data field? I think it uses the average speed since starting the navigation. 

When want to load a course and I go to a list of courses on the watch it shows estimated completion time for each course. 

These times are based on a extremely fast pace as you can see considering these are mountain courses. Furthermore, there is a white arrow during navigation that is always shown initially when the course is started. It seems to be moving on the course with the pace according to that estimated time. I suppose that is some kind of virtual racer even though I never use Pace Pro, never specify target time or pace for the course, or anything like that.

That must be from the time values in the file. In plotaroute.com and other sites there are different options for calculating time. In GC I see a simple "Pace calculator (virtual partner)". Never tried that, but as it say virtual partner, I suppose it gives you time values in file.

These timestamps are saved in the course FIT file by Garmin Connect. There isn't any Pace pro or virtual pacer or pace specified by me, or anything like that. Garmin Connect just inserts some pace data based on something, probably my marathon pace or something like that.

For example, I've just looked at a course FIT for my last 100K race. I can see that there are timestamps where the start time corresponds to the time when I created the course and the finish time is 8 hours from the start time. In reality, I was hoping to finish it in 14 hours and ended up finishing it in 17 hours. This course was imported from a GPX file which doesn't have timestamps, which means these timestamps were inserted by Garmin Connect.

The 8 hour finish time looks to be based on my predicted marathon finish time, which is a ridiculous assumption for a difficult 100K race with 12,500 ft of elevation gain. I don't think anyone had finished it in 8 hours.

One reason it is annoying is that these timestamps seem to control the "virtual pacer" or whatever it is called - the gray arrow that appears on the course that I am supposed to keep up with. But in reality that virtual pacer always moves much faster than me, so it is completely useless.

#Matthias# said:

You mean the ETA data field? I think it uses the average speed since starting the navigation. 

In my experience it's not as simple as that. It seems to me that the navigation ETA field (and the estimated finish time when you set a distance target, such as 10k or half marathon) looks at your average speed for a recent period of time (e.g. last 5-10 minutes), not for the entire activity.

I see this if I speed up or slow down significantly during an activity with navigation or a distance target.

silentvoyager said:

The 8 hour finish time looks to be based on my predicted marathon finish time, which is a ridiculous assumption for a difficult 100K race with 12,500 ft of elevation gain. I don't think anyone had finished it in 8 hours.

One reason it is annoying is that these timestamps seem to control the "virtual pacer" or whatever it is called - the gray arrow that appears on the course that I am supposed to keep up with. But in reality that virtual pacer always moves much faster than me, so it is completely useless.

Dumb q but is it an option for you to set the the expected pace manually when you create the course?

Also, if you've run the route before, can you create a course from the previous activity so it uses the previous pace?

I realize that doesn't really address your main concern (that the estimates suck).

There is already an interesting thread in the fenix 6 forum

Course ETE / ETA - what's it based on?

https://forums.garmin.com/outdoor-recreation/outdoor-recreation/f/fenix-6-series/262898/course-ete-eta---what-s-it-based-on