.
Have the points in Stephen Senn’s guest post fully come across? Responding to comments from diverse directions has given Senn a lot of work, for which I’m very grateful. But I say we should not leave off the topic just yet. I don’t think the core of Senn’s argument has gotten the attention it deserves. So, we’re not done yet.[0]
I will write my commentary in two parts, so please return for Part II. In Part I, I’ll attempt to give an overarching version of Senn’s warning (“Be careful what you wish for”) and his main recommendation. He will tell me if he disagrees. All quotes are from his post. In Senn’s opening paragraph:
…Even if a hypothesis is rejected and the effect is assumed genuine, it does not mean it is important…many a distinguished commentator on clinical trials has confused the difference you would be happy to find with the difference you would not like to miss. The former is smaller than the latter. For reasons I have explained in this blog [reblogged here], you should use the latter for determining the sample size as part of a conventional power calculation.
A seminal controversy in statistical inference is whether error probabilities associated with an inference method are evidentially relevant once the data are in hand. Frequentist error statisticians say yes; Bayesians say no. A “no” answer goes hand in hand with holding the Likelihood Principle (LP), which follows from inference by Bayes theorem. A “yes” answer violates the LP (also called the strong LP). The reason error probabilities drop out according to the LP is that it follows from the LP that all the evidence from the data is contained in the likelihood ratios (at least for inference within a statistical model). For the error statistician, likelihood ratios are merely measures of comparative fit, and omit crucial information about their reliability. A dramatic illustration of this disagreement involves optional stopping, and it’s the one to which Roderick Little turns in the chapter “Do you like the likelihood principle?” in 
Around a year ago, Professor Rod Little asked me if I’d mind being on the cover of a book he was finishing along with Fisher, Neyman and some others (can you identify the others?). Mind? The book is 
Error Statistics Philosophy 