You are over thinking it but not by too much.

If you have a voltage regulator on board that can only deliver 10A then 99.999% of the time you be fine until every servo wants full current at the same time. You will then be relying on the output capacitors in the regulator, and any filter capacitors in the servos and rx to make up the current difference. The one saving grace is that as the voltage drops under load, the amount of current the servos can draw will reduce, so while they may draw 16A in total at say 6V (?) that 16A peak might just drop back to 10A as the voltage sags below 4V.

All of the above can happen in the blink of an eye and you'll need an oscilloscope to spot it, but one microcontroller I used to work with would intitate a brownout reset if the supply voltage dropped to ~2V for longer than 100us (0.0001 seconds).

In practice this means that it's highly unlikely a Li-Fe pack will get pulled down to a low enough voltage for a long enough time to cause a problem, but apparently in the early days of 2.4GHz, a nearly flat 4-cell Ni-Cd pack + digital servos didn't leave you with much voltage overhead and brownout's did happen. I tend to size regulators for about 1/2 of worst case current draw and battery capacity enough for about 5 safe flights.

I have a 10A reg in one plane (6 brushless digitals of various sizes) with a 1.3Ah Li-Po for example and have flown the same plane for 1 flight with a 450Mah Li-Po.

As a side note, a servo that is stationary but unloaded will draw "stall" current until it gets moving, meaning the current peaks a servo is drawing while humming away in the pits is the full stall current but only for a fraction of a second. It's when those fractions of a second all happen at the same time that something bad can happen, but it's very rare. I watched the current draw of my two elevator servos as they buzzed away under the weight of the surface and there were two little 1.7A hills on the trace which joined up and become a 3.4A mountain, then split back in to two little 1.7A hills as the 300hz control frequency in each servo must have been very slightly different. Fun to watch if you have nothing better to do??