PCV won't change the pumping loss because it is a combustible mixture. PCV flow must be accompanied by reduced flow through the carb. EGR does change the pumping work loss because it is inert. Adding inert exhaust gas does not change the amount of new mixture through the carb so you get a net increase in intake manifold pressure. Increased intake pressure translates directly to reduced pumping loss.
Evacuating the crankcase doesn't help because you pay an offsetting penalty on the exhaust stroke. We've been talking about intake stroke because the intake is throttled but the pumping work argument applies to the exhaust stroke as well. Pumping work on the intake stroke is a loss proportional to the difference between crankcase pressure and intake manifold pressure. Likewise exhaust pumping work is proportional to the difference between exhaust manifold pressure and crankcase pressure. Fortunately the exhaust manifold isn't throttled so this loss is small. If you evacuate the crankcase it helps intake pumping but hurts exhaust pumping in an equal trade off.
Fig2 on this link can be used to understand all this discussion and more:
http://www.mechadyne-int.com/vva-reference/part-load-pumping-losses-si-engineSupercharging, turbocharging, valve event timing, pretty much every aspect of engine operation affects the shape of this P-V diagram. It's just a tool to understand what's going on inside the engine.
For the descending piston A versus B yes, you are correct. This is why the crankcase pressure is roughly constant at 15psi or one atmosphere absolute pressure. Crankcase gases pushing from under piston A to piston B and spinning crankshaft drag are additional losses above and beyond the pumping loss I'm talking about.
