<< Similarly if there are 10 worthless drugs that enter a phase ii for every one that is actually worthwhile then a p of 0.04 implies not a 4% chance that the drug is worthless, but actually much higher (I haven't specified the chance of a false negative, so I can't calculate an exact number).>>
Many thanks. Your post helps. The `bowler' example got me to think that poster DewDilligence was refering to different phenomenon altogether.
To paraphrase, I think what you are saying is that there is always a chance that stat-sig results we see in a Phase 2 trial are coming from a (truly) ineffective drug (because of limitations inherent in any trial, etc.). And therefore no one should expect positive results from the subsequent Phase 3 trial to be a shoe in even with highly stat-sig results from the Phase 2 trial.
I don't think anyone would have any trouble understanding that (if I am understanding your post correctly :-)) just as I also don't think anyone (or very few I hope) believes that 9902b has a ~95% (or some such ballpark figure based on the p value from 9901) chance of success. The current stock price and the fact we see anxiety about the pending results of 9902b suggest that people are aware of this issue (even if they are not using so many words).