Yeah qusp's suggestion was stupid ... because scientific approach to electronics is insanity and never produced a good product ever. Right.
ALL that qusp/tintin/others and myself have said is that it may well be a waste of time to go into this project without having some basic science behind the component and design choices and an understanding of the impact of changing one choice may have on the rest of the design.
Arbitrarily deciding to add ketchup to a meal won't kill you but it's not going to impress. However, if you and your friends like ketchup, then based on your only available anecdotal evidence you might think your creation is the best meal ever.