Yep, in broad strokes, randomisation removes any bias or confounding effects such as age, gender, behaviour, genetic disposition etc. The price you pay is in pr

Yep, in broad strokes, randomisation removes any bias or confounding effects such as age, gender, behaviour, genetic disposition etc. The price you pay is in precision; randomisation and the best practice analysis (intention to treat) tend to produce weaker signals of effectiveness compared to the control treatment. But that price is worth paying because randomisation is really the only way to be sure that you are seeing a real effect caused by your treatment.