In any discrete instance it may seem like that, but what you interpret as "strange ideas and dogma that falls apart when tested in real world situations" to a p

In any discrete instance it may seem like that, but what you interpret as "strange ideas and dogma that falls apart when tested in real world situations" to a practitioner is a history of different approaches to a problem that have been tried in the past, a few standard ones that are sort of your go-to, and a huge amount of case studies and research about why different approaches worked or not and what might make it work or not again in the future. The problem always is that it very difficult for an observer to tell the difference between an educated guess and just a, like, regular guess. So I don't fault anyone for interpreting it like that.