Builders usually approach their craft through one of two avenues - by following tradition or via the scientific method. I have one foot firmly in each camp (for what its worth, my head is also in the clouds, and one hand is in the cookie jar).

Steel string guitars have changed very little since C.F. Martin's first X-braced guitars of the 1850s. Traditional design has proved far more successful than most of the "improvements" we've seen through the decades.

On the other hand, I am an engineer and I think science plays an important role too. Science has failed (so far) to characterize the complexities of the whole guitar, but it is worthwhile to measure stiffness and density of the wood and the frequency response of a soundboard under construction. By correlating these objective measurements with the subjective sound of the finished guitar, I can get the desired result more easily than when relying strictly on ear and experience.