These are entirely possible but would rely on a very complex event chain. The dev team is planning on making a number of changes to the editor, so I've held off on anything too crazy until then.
I've created and shared an alternative history scenario along some of these lines you describe (see URL in my signature)
agree its perfectly doable and interesting, but for a formal release there are a few issues:
a) its incredibly hard to test the consequences of branching event chains. Sometimes a low prob outcome can simply be game ending.
b) testing each branch so it works technically is harder than might appear
c) in combination to include contingency/branching events would have added a lot of time to the development of the game.
I did a lot of work for AGEOD's Pride of Nations, primarily cleaning up the event chains, adding some neat chrome and trying to remove anything that stopped them from working. Even for (b), it took a year (unpaid etc and that was using ideas and material generated by the wider community) to get what became the final patch out. The safety net in PoN to (a) was the extent it was scripted and that you couldn't go too far off track (it wasn't a world conquest game and it had at its core the idea that the main European nation states were fixed), but even so.
Further comment, the later you get the more these problems compound. Also when you come to massive events like the Russian Revolution you need watch out that interesting modelling doesn't create very unwelcome game play.
So lets say you want a fun, Soviet collapse in November 1941 event. Add your conditions and you can guarentee that some axis players will simply aim at them, now it might be realistic, you might assign a low probability etc but ...
Yes, and things start to go off track because the AI has a tough time with certain aspects of the game... For example, opening up Turkey for combat isn't really that much fun because it has a hard time handling it.