Basicamente, in fighter aircraft you need a lot more of elevator than rudder.
For most aircraft, the required size of the elevator is determined by one of the two corner cases:
- Sufficient control and balance at low speed, particularly on landing;
- Sufficient control for high-G manoeuvres.
For fighters and aerobatic aircraft, the latter is usually the main factor; for others, the former.
In contrast, rudder is usually sized for crosswind landing and, if applicable, for the asymmetric thrust (one engine out) case. It's easy to see that due to closely packed engines on most fighters and higher speeds, both these factors are relatively smaller compared to other aircraft (e.g. airliners).(1) Whereas the elevator demand gets worse. This is why.
There are two main problems that arise at supersonic speeds.
The centre of pressure of the wing (and the neutral point) moves back significantly. This obviously creates a strong nose-down moment (and makes the aircraft statically very stable in pitch for the same CG), requiring a lot of control to re-trim the aircraft and execute any manoeuvre. This is the reason for 'Mach tuck' (in fact, this is the Mach tuck), and not the rubbish from the quoted Wikipedia passage.
-
Efficiency of the elevator fades. At lower subsonic speeds, the elevator effects the air not only above/below it, but also ahead of it. The approximate efficiency of the conventional elevator at such speeds can be expressed as
$$k_{\mathrm{elev}} = \sqrt{\frac{ S_{\mathrm{elev}}}{S_{\mathrm{stab}}}}$$
Therefore if the elevator $S_{\mathrm{elev}}$ occupies half of the stabiliser area $S_{\mathrm{stab}}$, its effect will be not $ $ 0.5 mas sim $\sqrt{0.5} \approx 0.7$ (compared to the all-moving surface). But at supersonic speeds such propagation upstream cannot happen, and the efficiency falls to the area ratio, i.e. $ $ 0.5 in this example. This favous all-moving surfaces.
The second point affects the rudder just the same, of course. But the static directional stability (point 1) doesn't increase like it happens with pitch (in fact, it can fall), and the overall demand is lower, so conventional rudder is still acceptable. Nevertheless, there are a few aircraft with all-moving vertical tail, for example, Tu-160.
In other circumstances, conventional control surfaces are better (i.e. lighter), so they are used whenever possible.
(1) This should not be confused with the whole vertical tail sizing. This one is often determined by the stability dynamics, and directional stability of supersonic aircraft can be rather bad, demanding big tail - but not necessarily big rudder.
