Obviously, probability theory has it's place. But it seems like over the past hundred, maybe even 30 years, we've let p minimization and r2 maximization become a substitute for actually having sufficient sample sizes, proper controls, replication, and quality study design.

What we have now is mass produced research that doesn't really tell us anything. The very numbers of stats fundamentalists indicate that the MAJORITY of published research is wrong. Just a thought, but maybe it's because we've replaced actual analysis with a bunch of "p-value" hokum.