Thanks Nate. I really only did the most standard frequentist analysis for the task at hand. I’d be happy if you can update the article so it presents the frequentist position more fairly.

I have nothing against computation intensity, it is often inevitable for more complex tasks. I just don’t like to go there when not necessary.

My question for Bayesians is what happens with the difference in interpretation when faced with the all too common situation of a Bayesian and frequentist test resulting in the same output (major differences seem to enter only when there is sequential analysis of data). If the p-value is the same as the posterior probability numerically, does it mean it is in fact a posterior probability in cases of supposed lack of prior information about the parameter of interest? Or is it not the right posterior probability as it is the Bayesian analysis that is incorrect in using minimally-informative priors?