This is also one of the most basic lessons of decision making: decisions are independent of the results of those decisions. The results are inevitably shaped by luck too, and it's impossible to say which one caused the result. You can, of course evaluate a decision as a good one if its correlated to a good result over a high number of trials, but survivor bias and success stories deal with very few trials.
(Please correct me if I'm incorrectly tying decision-making and correlation vs. causation into this)
The crux of the matter though, is that we have no alternatives. We don't have the ability to perform controlled trials to verify any of our theories like in science, so we must look at the past and derive any lessons we can.
The crux of the matter though, is that we have no alternatives. We don't have the ability to perform controlled trials to verify any of our theories like in science, so we must look at the past and derive any lessons we can.