The US is a common law nation. We replace bits and pieces of the common law with legislation but when you strip all that away, the common law is what is left. I'm a great believer in primum non nocere, "first do no harm". So when I look at politics, at government, the first thing to ask is whether the current state of affairs created by legislation is better or worse than the common law. If it is worse, then we would be improving things by stripping out the law and returning to the status quo ante of the common law, even if the common law has not been enforced in living memory.
This leads me to ask a lot of simple questions and to hope that there were a way to automate and flag when laws actually net out to a worse result than what would happen by readopting the common law and using modern tools to enforce it. This is the sort of thing that would ideally be measured in an automated fashion to get a first order approximation and then where the results are close, a more detailed examination would proceed.
For example, how bad do the results of a police force have to be before restoring the original 'hue and cry' system would be an improvement? Are there any actual jurisdictions where this situation is a reality? How good would a modern hue and cry system be? How would such a system affect community, impede or enhance commerce, or provide justice? We don't know because we're not asking the questions.
In the vast majority of cases, the answer is going to come back that statutory law is an improvement but merely by routinely asking the question, you set a floor on performance, a standard by which government is not allowed to go below without pretty drastic consequences. And that's not a bad thing to have.