However, in my latest tests with official 1.01 version I have seen some really weird battle outcomes which cannot be explained by above argument. Two weeks ago, I wrote about this in another thread but nobody from official team neither confirmed this as a bug or give reasonable explanation about why some clearly won battles are declared as defeats. I will repeat here what I wrote before.

In above tables you can see battle outcomes where two legions (both have same number of men and horses) fight each other. Only difference among three tests is related to who is in command of Populares legion). Out of 33 battles, in only 4 battles Populares are credited with victory (all four in situation where Marius was in command). However, if you pay attention to numbers, it is really difficult to find explanation why Populares weren't victorious in much more battles. For easy reference, I circled in red these disputed battles.
For example, take a look at 11th battle in Marius test (third table, last battle). In this battle Optimates lost badly. They suffered [color="#FFFF00"]4.797 infantry casualties[/color] while Populares suffered [color="#FFFF00"]only 2.706[/color]. That is a huge difference! Also, Optimates lost entire cavalry while Populares didn't lose a single horseman. By all standards this is a clear Populares victory but still, they retreated and Optimates were declared victorious.
As you may see, there are other battles with similar results which indicates this is not a single exception but rather systematic phenomenon. Do I have to mention what effects this may have when someone is playing against human (or even AI) opponent?
My questions are:
1.) Is this a bug or WAD feature?
2.) If it is a bug, are devs aware of it and when we can expect to see it addressed?
3.) If it is a WAD (which I doubt), can someone give reasonable explanation about above disputed battle outcomes?