Will, you're confusing social science and hard science. There are hard truths in hard science. But humans are so complex and variable that they're notoriously hard to predict. As for "verifiable better ways to do things" - who gets to decide what is "better?" "Better" for whom? By what standard?
That's why - in my humble opinion - going with what makes sense is not sensible, because what makes sense to one person won't make sense to another. The question should be what works, not what makes sense. I have written about that in the past. If you're interested,
click here. But be forewarned - it's typically long-winded lawyer talk.