Some people may remember my .913, .915, and .952 playthroughs. I am considering a new one, and I am doing it to test a theory.
The theory is as follows:
In .98, overpowered heroes are the most effective strategy, and the productivity/research costs and facilities make training units nonviable.
As any wanna-be-scientist (I am an engineer, and everyone knows we are just people who did not make it as scientists) I am going to devise a test to falsify the theory.
So this is what I plan to do: I will make a custom race and a custom sovereign that will be exclusively focused on building the best trained units possible. Every single point of design will go in supporting research/production/units.
Then I will play the game to win, that is, I will do the best I can with what I have.
Then if I win due to trained units, it will mean that my theory is wrong. If I lose, or win due to my heroes , we'll look for another experiment.
Here is what I plan to do:
Race: No ranged weapons, vulnerable to magic, defensive, master scout, light plate, master smiths, warrior caste.
Sovereign: General, Earth, Life, Scholar, crown+sword.
Difficulty: If it's not insane, it's usually not much of a challenge, but I can be convinced on this one.
Anyway.
Please do not use this thread to argue whether the theory is correct. I would like to keep it for suggestion for the playthrough. Also, remember your scientific method. I could design a uberhero, and kick ass with him, but that won't prove anything. You can't prove a theory, you can only fail to disprove it.