Thank you for taking the time to answer in such great detail!
You described excactly my problem: the test profile is a trap and that is good, so I recieve proofs of every delivery.
That it is unfortunately also in the reports I found by filtering by domain, just as you described.
So, if e.g. the delivery goes to 2 profiles and one is the trap test profile, the opening rate is shown as 50% when the test delivery was opened but not the real recipient. This messes up the statistics from a customers perspective if the delivery is on a small scale and at least reduces trust in the accuracy of the numbers in the report, as test profiles shouldn't be in the reports.
So, is there any way to filter some profiles from the reports afterwards?
I didn't find any way to enter exclusion or filtering rules like "NOT domain = xyz.com".