Can an argument be rational if it is not founded on evidence? I don't think it is rational to adopt beliefs as a grounds for reasoning without any reasons for adopting them.

Gravity is not, in fact, the attraction of lesser mass to greater mass. Gravity is the repulsion of lesser mass by greater mass.

In the case of the Earth, this means you are repulsed toward the Earth by the weight of the column of space "above" you, and are only prevented from falling through the Earth by its density. The Earth is relatively equally displaced by the column surrounding it, meaning you do not notice this effect.

Gravity seems to come in waves because all things in the relevant column are moving, but the shifting of mass is almost unimportant to your perception because it is such a small value. The magnitude and frequency is such that it goes unnoticed to all but the most sensitive equipment.

.... completely no evidence available, but, it sounds kind of reasonable.

You actually cited evidence in your example:

>prevented from falling through the Earth

>Gravity seems to come in waves

>the shifting of mass is almost unimportant to your perception [...] unnoticed to all but the most sensitive equipment.

Point taken. With that in mind, yes, you're left with Shrodinger's cat.

You can't find evidence for everything in the moment. If you're forced to act it makes sense to trust what seems most rational given limited knowledge.

It doesn't make sense when you don't need an answer at that time. If you can find evidence, you should.

Depends on who you ask – there's a certain school of economists that insists that a sufficiently "rational" argument is in fact preferable to evidence.