No amount of math can tease apart the indiscernibly complex. If there's any structure at all, you can find it with enough math and enough data; that's exactly

No amount of math can tease apart the indiscernibly complex. If there's any structure at all, you can find it with enough math and enough data; that's exactly the purpose of statistics. But without controls, and without repeatable experimentation, we can not have a true scientific method. The scientific method requires hypothesis, predictions, and repeatable experimentations. No, this misunderstands science. All the scientific method requires is testable predictions. Predicting the outcome of a controlled experiment is merely a very efficient way to go about it. There are plenty of scientific fields where simple controlled experiments aren't really viable, and people get by. It's just more difficult, and requires more discipline from researchers.