I have this very simple model whereby there is an altitude below which all voids in rock are water filled.
That holds good in rocks with a high primary porosity, such as sandstones but not in limestones where there is a very low primary porosity but a high secondary (fracture) porosity. When water flow is highly localised within the rock, as it is confined to fractures, that model does not stand up well. There is no mechanism whereby head (water pressure) can be equalised "sideways" between flow routes in independent voids and local geologic features, such as the Derbyshire clay wayboards, will have a more marked effect.
Of course on a regional scale, there will be a degree of corroboration, given that many springs will be a similar heights, at the base of limestone outcrops for example, and sinks will be at similar heights, at the top of a limestone outcrop, but that could almost be described as coincidence rather than anything else. I am thinking here of the geologically relatively simple Yorkshire dales.
To take the example I know best (and which has probably been most intensively studied in these respects) of Mendip, if the "water table" was a real concept and drainage was thus simply controlled by hydraulic gradient, then the flow pattern on Eastern Mendip shown on
page 7 of this paper by Dave Drew could not exist. This was the first example of flows crossing without mixing. Other examples include the flow from swallets in the Hillgrove area which reaches Rodney Stoke Rising having crossed, without mixing, flow to Wookey Hole. Probably the best example is Wigmore, where a group of swallets "should" send water to Wookey, or even north towards Sherborne, were it controlled by the steepest available hydraulic gradient, but instead the water flows all the way to Cheddar. This is probably a relatively old flow route, given that the limestone at Cheddar was stripped of it impermeable overburden much earlier than the flanks of the Hill further east, allowing springs to develop. However, the important point, here, is that even though those steeper flow routes became available later
they were not used. This is because there is not a simple water table that acts in a simple fashion in limestone.
As always the devil is in the detail.