1. We can measure things that we cannot hear.
2. Some of the things we can measure matter more than others.
So I can reach a few conclusions from this:
1. If I cannot measure a difference, then I certainly cannot hear a difference.
2. If I can measure a difference, then I may or may not be able to hear a difference.
3. To tell how audible a measurable difference makes, one can employ subjective blind listening tests. That will help determine what is audible, and by how much.
I strongly agree with your point 3 that a properly controlled blind test will go a long way in bridging the gap between perception and measurement.
Non blinded tests are often biased as a result of expectations arsing from visual cues, prices of equipments and even equipment brand choices
I am not saying that blind tests are the ultimate method here, but it is a far better option than the visual testing method.