Maybe I didn't show the difference clearly enough. On one side you have an assumption on how things work and you just check if they apply and assume if they don

Maybe I didn't show the difference clearly enough. On one side you have an assumption on how things work and you just check if they apply and assume if they don't apply then the negative must apply. But often there are more options we simply don't know about. So if your assumption A is false it doesn't mean it must be "not A -> B", it can as well be C which you don't know exists as well. And if your assumptions proofs correctly it can actually be because of D, which looks quite similar to your assumption A. You are absolutely right that we can not avoid completely to fall into this fallacy. But we can try to mimize its effect by not asking "A or not A?" but instead ask "What is that?" and "How to describe that?".