Mon Sep 14, 2015 5:04 am
You can get a sense of the magnitude of defeat/victory from the NM adjustment after the battle (this can be a little wonky, too, but is justified for the most part; NM adjustments reflect how your country's citizens respond to the battle, so it is not infallible). As for odd battle results, usually the side that has the greater losses loses the battle, but not always. Sometimes there are pyrrhic victories, which is as it should be, in my opinion. Sometimes in the chaos of war the side that has taken less damage still retreats. If you turn on the error logging you can dig into the log file after a battle to try to understand more of what happened. (You might want to focus on a small battle, to not get overwhelmed. There are a lot of metrics to take into account for each battle.) Or you can sometimes piece together what went wrong by the reports from the battle results, such as the number of failed morale checks by individual units.
Even with unpredictable battle results, if you do the right sorts of things a general should be doing going into battle, you will tend to get the battle results that you want to see most often. I have always liked how the Athena battle engine is an unpredictable black box that still rewards good military behavior.