Question

When designing an experiment how do you choose a system to investigate

Answer 1

you pick a system for which no control sample exists, so that no one can show that the alleged causal relationships you assert do not, in fact, lead to the phenomenon you claim to have observed.