I say leave the randomness untouched. Even or close battles are gambles, and gambles always make games interesting . Either way, it's just bad strategy to bet everything on one single even fight. You deserve to lose if you do that
First I'd like to say that what a marvelous simple game this is.
So I was wondering about planet upgrades. All the upgrades seem to increase the planet production by 3, no matter what the level is. By my simple logic, isn't it more cost effective to always upgrade the planets that have the lowest production? Like if you upgrade a planet with 12 production to 15, it takes 4 turns to make up for the lost ships for the turn that the planet produces nothing. Now if I upgrade a planet with 4 production, I've already profited by 2 ships on second turn. Is my logic correct? The AI seems to like to develop its homeworld, isn't it losing ships that way? The highest difficulty feels a little too easy, because I haven't even played the game much and I almost never lose. This might be one reason. It also really hurts your economy if you lose highly upgraded planets. I've came back from desperate situations countless times by taking a highly upgraded planet from the AI while they only get a level 2 planet in return.
Planet size only matters in the beginning, and if factory updates are off. One thing that might mix it up a little would be that the production update increase is tied to the planet size. Planets with the starting production of 4 would increase by 3, 6 would increase by 4 and 8 would increase by 5. That way planet size would actually continue to matter more throughout the game, though I have no idea what implications that would have on the game balance.
EDIT: Now that I read the map making instructions, seems like you can set the amount that planet upgrades increase the production. Maybe I should test that by making a map of my own.