[Adam Y.]

Perhaps what I learned in my engineering design course could help. The trick is first to think of the factors there are. Doesn't matter what goals they are as long as they are factors like cost, durability, ability to throw poofy balls in top goal, speed, or time to build. You next have to assign a number of a scale of one to ten to represent the significane of that attribute in the design with a one being the least important and a ten being the most. Next you come up with a bunch of designs. Rate each design based on the preceeding factors on the same scale based on how well it would accomplish those goals. Multiply the factors by how well each design accomplishes the goals and then add the numbers you get to a total score. The one with the highest score should be the one you should pick. Still pretty subjective but at least it's a way to organize the subjectivity.