this is a method, you are free to like or dislike. Better than all of your useless methods. Actually you have no method. Did somebody make a working like that stay here? I don't remember, sorry. For make a critism, you should claim a better one firstly.
Your "method" is, objectively, not better than actual statistical models, scientific methodology, or even the Zetetic Method that you guys like so much around here. Your "method" consists of forming a conclusion that you want, finding any hodge podge of "data" that support it, using assumptions and imagination to cobble them all together, and then proclaiming the result that you engineered (badly, I might add) to be the "truth".
My better suggestion would be to use observable facts to confirm or deny hypotheses, controlling for variables so you can correctly isolate the true causes of effects. Limit assumptions whenever possible and claim them openly when they are unavoidable so everyone can see what your argument is hinged on, and most importantly...be open to the possibility that your idea is incorrect, no matter how badly you want it to be true.