No, you make a hypothesis, test it, and draw a conclusion. You then create a theory based on the evidence, which is a type of assumption. The main point I am making is that they go about the scientific method in reverse. You do not make a theory and then find the evidence for it.
You have no idea what you're talking about either. You cannot conduct an experiment without first taking a guess at what's going on, since that initial guess will tell you a great deal about what kind of experiment is needed to produce evidence that will challenge your guess. Going in and experimenting first makes little sense, what are you going to run an experiment on? You'll have to sit down and think about it either way.
Also broad physical phenomena is not scientific evidence. Perfect evidence about something needs no possible outside influence, hence controlled conditions. An experiment is conducted to collect data under these controlled conditions, and then the data goes back to either support your idea (which is boring, nothing new) or challenge it (which means the game isn't over yet and you still have a job).
