>> somewhere between 32 mo and 38 months or so. So I think this is very reliable. <<
NO disagreement there - I was referring to control median of 25-month. The bulls better hope the MST is less than 30-month.
>>A typical Futility Test is to halt the trial if the power is under 20 (or 25 or some equally low power) percent assuming the same ASSUMED HR that was used in the original powering assumption. That means at 50% info fraction a trial won't be halted for futility unless it is actually has a neutral HR (1.00) or is actually the wrong way.<<
Wrong. I would say the HR is not much worse than 0.9 at the interim already, at 20-25% conditional power. Martin says 5% chance for second and final, he essentially is claiming the true hazard ratio is 1 or at the null hypothesis, when the overall power will equal to the significance level remaining at the second and final analysis. [edit: rephrase type I error]
Market Data 
Markets 
AI
