I think that the game designers have a vision of how they expect the game to work, of how the player should behave to achieve victory. They set up the rules of the game so that it is possible for someone to play the way they expect. They don’t check if the rules they specified allow for really crazy, extreme solutions.
Now, in some situations it’s just way too difficult. Certain very creative strategies just can’t be foreseen. But I very often find that designers don’t even undertake a fairly cursory examination of the solution space. I think Galciv3 definitely falls into this category. “What happens if I try to maximum this planet’s (industry|research|income)?”, “What happens if I try to maximize a ship’s sensor range?”, “What happens if I try to maximize a ship’s speed?”, “What happens if I try to maximize fighters?”, “What happens if I try to minimize weapon reload time?”, “What if I try to trade stuff I don’t need to supplement my income?”. The answers to each of these questions seem to shock the developers, but they really aren’t crazy, off-the-wall things to try.
I think a lot comes down to the basic mathematical formulas used in the game. Anyone who has played a lot of these games can sniff out increasing marginal returns, and knows the types of formulas that are likely to explode if you are allowed to push a variable too far. Which is not to say that everything should be boring. You want powerful synergies to exist. But the designer needs to know what synergies they are putting in, needs to explore the limits thereof. Frankly I wonder if it’s not possible to have some sort of math analysis package help with this “here are the formulas, find all the feasible local maxima and minima.”
