But a lot of science relies upon those what's more likely?
For example what's more likely a GC solar system with epicycles and so on, or a HC solar system?
I don't know that I would agree with that comparison. Two theories supported by evidence can be compared with respect to the quality of that evidence - what are the predictions made, how much goes unexplained, how much is assumed? Isotropy is an assumption. Comparing an assumption to a hypothesis doesn't carry with it the same easy comparison.
Like, even beyond the fact isotropy is questioned, I wouldn't necessarily count it as even a core assumption. If she was coming for homogeneity, I think you'd have more of an argument as 'The laws of physics can change depending on time/location' would kill any attempt to derive an empirical system when different experiments for the same thing might have different causes, etc - but even then, that doesn't make it true, and they are ideals that have been criticised. Plus, so long as the violations are constant or predictable, science remains possible so there seem to be minimal negative consequences, and assumptions carry no evidential weight to me.
And yes, a FE can come with its own ramifications, and its own what's more likely question.
What's more likely, that there is a universal down, and in addition to that we also have something causing things like the cavendish experiment, and something causing the sun to spiral around, and bendy light to make things disappear from the bottom up, and so on. Or is Earth round, with gravity causing things to fall, Earth to orbit the sun, and so on, with light travelling mostly in a straight line with the curvature of Earth causing objects to disappear from the bottom up.
My stance is always that even if we grant for the purposes of argument that the Earth is flat, it's never going to replace RET in our lifetimes because there's simply too much for it to do. Total theory comparison is just never going to be feasible, hence my willingness to focus on it piece by piece.
Like, if we find a fully working, mathematical model for what brings things down, for eclipses, for tides and weather patterns, for the horizon etc etc, you'd then have to bring them all under the same model, and then untangle conspiracy elements and find areas it works better and...
This comes down to what you consider to be the system. I would say the foam ball decreases the density but increases the size.
I'm viewing this more continuously - if we shift to gravity for a second, and we have a ball held above a table, while the gravitational interaction between the two is minimal, it exists - the table has an effect on the ball pulling the ball down, and the ball pulling the table up, albeit insignificantly. At the same time, the air has its even more minimal force - the air behind the ball is also acting on the table, the air in front of the ball acting on the ball in the direction of the table, etc.
Similarly, the air that would take the place of the ball in the jug-and-density scenario could still be modelled as part of the system - the air is sorting itself out with respect to density too. It just obviously isn't any denser than itself, so there's no need to incoporate it.
But taking it your way as well, it's fair to say that the density of the system is increased by more than its size, just by comparing the ball to the air, so that seems functional too.
But yes, the foam ball is somewhat superfluous, as there is still the question of where this downwards force comes from in the first place, and you can show the problem with an object less dense than steel sitting on the plate of a steel scale with steel springs.
Hence, probably best to wait for an explanation of the force in the first place.
I think a compressed spring makes sense, the way I said before. Rote comparison of the density of the elements involved seems like a too simplistic way of modelling this - after all, you can take a spring of solid steel, and a cube of the same amount of steel, just solid, and it's clear that the cube would be denser even if they're both steel. An object less dense than steel, can still end up more dense than a steel spring.
Ditto, a steel plate on a spring, the overall system of the plate and springs could hypothetically be lense dense than a material less dense than steel itself, just because there are gaps. Volume will always be a factor.
The issue I do see tripping this model up is the lack of a reaction force from an object at rest, so I understand you there, and that seems to be where the biggest problem arises, but I do think there are doors available - or at least avenues for experimentation.