Nearly all decisions are recoverable. Most things are rather low risk, and when there is significant risk there's typically a way to hedge that risk, like through insurance or contingencies. You'll make countless decisions in life and at most 1 might kill you.

In nearly all circumstances where you're actually fretting over the decision, you're choosing between two reasonable options. It's unlikely one option will be substantially more recoverable than the other, and if you're still considering it anyways you likely have a good reason.

I think the better way of evaluating tough decisions is to think about which choice leaves you in a better position to make the next choice. The real dilemma in almost every decision is the uncertainty, and typically you'll learn a lot more from trying one thing than the other. Oftentimes also choosing a particular order to try different options leaves more options open later on.

Further, I notice that a lot of people see dilemmas where they don't exist. Practically every week I'm in a meeting where there is a heated discussion of whether to do A or B when we have to do A regardless of whether or not we do B and almost invariably the question of whether or not to do B will be answered by doing A.