@Stabbey, you have written a few posts above that you tested a lonely trooper against an enemy army by doing autoresolve. Could you please make the same series of tests, but just enter those fights as a dragon now? Write down the losses of your enemy, average them and compare them to your test results with autoresolve.
If you find, that there is a huge discrepency between these two approaches (i.e., your enemy's losses against your lonely trooper are much smaller in the RTS approach compared to autoresolve, or vice versa), then there is indeed an issue with autoresolve. If, in contrast, the results are pretty much the same, then it is how autoresolve should work.