If I run a big clinical trial pitting a placebo against a useless drug, the odds of getting the same results from both are low: one is going to appear more effe

If I run a big clinical trial pitting a placebo against a useless drug, the odds of getting the same results from both are low: one is going to appear more effective than another. If I run a hundred of these kinds of trials, I'm virtually guaranteed to see a few cases where one appears much more effective than another. And if I test my worthless drug against an existing treatment that was approved through the same process, I may as well be testing against another placebo. I'm not suggesting all drugs are worthless; only that drug companies do not need to produce effective drugs to stay in business.