All those explaining correlation-is-not-causation to scientists for the 500th time should take at least one semester of "You are not as smart as you think, nor are you very original, or know anything about the copious amounts of statistics scientists have to endure today. You're also conveniently living in a world that apparently allows for cheap and fast intervention studies on rare phenomena with long lead times and no apparent ethical qualms of subjecting people to potential harmful interventions. Because only that could allow you to forgo the imperfect (yet historically highly successful) observational studies"
(-Studies)