We create a lot of stories and subdivide each story into tasks. Estimates are made on the tasks.

Because of this, all stories appear to be unestimated, even though we have much better estimates when you take the aggregate of the subtasks.

Should GH be taking this into account?