I fully agree. We need to get rid of p-values because they don't mean what people think they do. And simulation helps set a realistic baseline expectation (instead of the "null effect" in NHST).<p>Another important approach here is specification curve analysis, which also explores the plausible range of effects to be found [1].<p>[1] <a href="https://pubmed.ncbi.nlm.nih.gov/32719546/" rel="nofollow">https://pubmed.ncbi.nlm.nih.gov/32719546/</a>