In order to disprove something like the flat earth, you need to have a fairly in depth understanding of what exactly the FE model claims. Then, you find some observable phenomenon - the length of a shadow at a particular time of day, the angle of your dish while watching satellite tv, the bahvior of objects near the horizon - and explain why the FE explanation does not produce the results observed in "Real life".
Paradoxical's solstice experiment is a good example. He, understanding the basics of the FE model, calculated two things: the theoretical length of a shadow cast on the Summer Solstice if the earth were flat, and the theoretical length of a shadow cast on the Summer Solstice if the earth were round. He made no assertions about which set of values was correct, only what would be implied by certain measurements. He found these theoretical values through some basic trigonometry, and a methodology that, at first glance, did not seem to conflict with FE canon.
So this wednesday I'm going to go out, put a stick in the ground, and measure its shadow. If I get value A, then that will be a point for Flat Earth. If I get value B, then RE has the edge.
This is a good example of how you can operate within the bounds of FE theory to come up with some theoretical observation, and then illustrate that the observed phenomenon does not happen, at least not according to FE canon.
In short, you cannot disprove FE by proving RE is more plausable. If you wan't to disprove FE, you have to show it cannot work.