It's not so much a measure of flying ability but of a way to compare models of different sizes. If you have a 1/2A model and a giant sized model of the same design, just scaled up, that have a lot of flying habits in common then the cubic wing loading will likely compare quite closely between the two. However if you look at the normal area loading the 1/2A model will have a wing loading that is perhaps only 1/3 to 1/4 of the wing loading of the big model.

None of these methods is perfect by any means. But the cubic loading provides a closer comparison when looking at scaling a design.