What exactly do you mean here?
Discussion
The trial and error method works generally, even when people behave irrationality, if there is a feedback mechanism (successes rewarded, errors punished).
I’d say that would mostly work for simple questions: should I eat donuts all day every day to stay fit. What about complicated questions? Does the covid vaccine provide net benefit? How can someone even observe something like this when the don’t have their own random control group? And what do we do with the high likelihood that they are unfamiliar with statistics?
Well, the central question is net benefit for who exactly. I believe that the individual themselves is the most aware about what is better for themselves. If I want to eat donuts all day - that's fine. The same for vaccine. What if I want to take risk? Which risk is worse and how to estimate those risks - that's up to individual. And who said that statistics is the best answer? What if he wants to play risks? E.g. not take a vaccine and then sell his sperm for thousands of dollars. People are free to experiment with their lives. That's the best path to happiness - let person do what he thinks is best for him. Maybe he is wrong, maybe he is right. Nobody has the right to take away the right to make errors from him.
I’m not saying people can’t choose for themselves.
I’m just saying if people rely on their own data collection and analysis, well people are pretty horrible doing that for topics like health. We are aware of so many logical fallacies and biases that humans predictably fall for. For example, we rarely actually take into account the survivorship bias in health gossip we hear. Our confirmation bias, availability heuristic, etc… That kind of stuff takes a properly set up and repeated study.
Not sure what you mean by who said statistics is the best answer. Statistics “is the discipline that concerns the collection, organization, analysis, interpretation, and presentation of data” and like I’ve said above people tend to have poor intuitions on how to transform observations into conclusions.
But I’m only talking about people who desire to have reliable conclusions. Folks who don’t (though I think I’ve never met one) don’t need to worry about it.