statisticians weren't able to convince the academic world that arbitrarily setting a "statistical significance" threshold is a gross misuse of statistics

What would the proper use of statistics be? I believe it is to produce a Bayesian posterior probability distribution, but that wouldn't be the opinion of a frequentist.