It's hard to define "results" here, though, isn't it?

Example: I once encountered someone who just "knew" that the raises they were getting were proof that their religion was true. They were less successful, then they prayed for years like they were taught, then they were more successful. Results, right?

But when you look at the census data: between accumulation of human capital and just-plain inflation, the median retiree has seen their income go up by an order of magnitude over their life time. Getting raise after raise and watching your salary go up ten-fold was the default life experience, not proof of divine intervention.

So to see "results" you can't always just compare "before" vs. "after". Even if you see results of your epiphanies, does that mean the epiphanies get the credit, or might you have seen the same improvement from the same amount of life experience, epiphany-free? Or even if the epiphanies did help: which ones? If you're seeking out epiphanies, odds are you're not just testing out one new idea at a time then waiting to rigorously analyze the independent results of each.

To reuse my example: even though income increase was a lousy metric, on other evidence I would actually say that their religion was providing many adherents with "actual power", on net: buying the whole package wasn't nearly as beneficial as picking out the value from the dross would have been, but it was better than the also-mixed-bag of beliefs that most competing religions and secular ideologies offered.