You are making the right point but you are not presenting the most important argument. The essential problem with economics is that an economy fulfills all criterias of a complex system and therefore — by the definition of what makes a system complex — can never be described with simple mathematical models.
That doesn’t make classical and neoliberal economic models worthless though. They’re still fine to describe market dynamics and the interplay of an economy’s sectors to first semester students. They’re also important when it comes to defining terminology and explaining essential economical ideas.
Also, especially in the world of micro economics, there are some universal thruths that are absolutely observable. For example that economies of scale almost certainly lead to a oligopoly or monopoly. Especially when they benefit from a network effect. See social networks, airlines, telco etc.
But what too many people in economics fail to realize: one cannot use simplified mathematical models to make predictions for overall economical developments nor to give political advice nor to even just explain a crisis in retrospect. Only models that are based on complexity theory can actually do that — and this is specifically hard in economics.
Creating models for complex systems involves the creation of rule sets for the decision-making process of a huge amount of individual actors and to then simulate their interactions. It’s the same kind of research that climate scientists use. Just that in the system that they research, the rule set for their actors is comparably stable. But the rule set in economical models must be valid for billions of individual actors and hundreds- of thousands of institutions. Not only are those rules hard to describe in mathematical terms but even worse — those rules are highly situational.
People make economical decisions by entirely different rules when poor or rich, save or in danger, in panic or confident. Dictatorships have different rationalities than democracies. Technology advances all the time in unpredictable speed and reshapes markets faster and faster. With nowadays knowledge and technology it’s basically impossible to find a model that can describe the world’s economical developments today and isn’t total nonsense within 5 years. The underlying rules for economical decision-making on the actor level is simply too volatile.
However, there is some really good stuff in economics with lots of substance. Take theories about market failure for example. It’s a well established and yet well ignored domain of economics. Mostly because it doesn’t fit into many people’s believes that free markets can regulate everything. And that’s where I’m with you. Economists are way too religious in their views and ignore obvious evidence when it doesn’t fit in their picture of the world.