The 19th century in the US is probably the largest data set when it comes to extremely low regulation markets. Hong Kong is a more recent example. The point being, there is a lot of actual real world data on each side of the fence.

> The point is that we can't take a market with N regulations, see what happens with (N-1) regulations being followed, and then say "see, that's what would happen without any regulations!"

That seems like the sort of evidence that's actually useful when making policy decisions about whether we should increase or decrease regulation. Abolishing all regulations is not practical, so what political theorists think would happen in a complete absence of regulation is irrelevant to any real-world issues.