[Spivak]

You basically can't do science like this and prove anything useful outside of group A is the more effective than group B. And that's assuming your test subjects can't tell the difference. That would tell you that at minimum either A or B has a non-zero effect because they both can't do nothing and be different. But proving that A is an improvement over not A is the important bit. A and B could both be worse than nothing but A is just less bad.

"find something that has a similar trip but does nothing" is actually equivalent to the original problem.