Simple way to test it--build the tinies and enough constructors to make the starbases at the end of a game, save, finish, and look at your military score. Then reload the save and build the starbases, finish again, and look at your score.
Oh, I was under the impression that your final score was determined by some sort of average of your scores every turn of the game (in which case what you do on the very last turn can't make a real difference no matter what), rather than just being determined by what you had at the end. Okay, I'll run an experiment like this.
I ask because I spent a bit of time figuring out really cost-effective ways to achieve military dominance, and it turns out that relying on ~5 fast medium ships to kill everything (thanks to utter tech dominance) results in a terrible military score, so now I need to figure out a different approach for metaverse games.