Ok, some thoughts on this:
1. Cataclysms.... not a bad idea, actually. It just depends on how they're done. What would be ultra frustrating would be to, you know, be going along just fine.... and then suddenly a corner of the map explodes into a horde of shrieking laser chainsaw skeletons and Minecraft creepers and then BAM the whole thing ends because RNG. And yet, the cataclysms need to still be DANGEROUS. Part of the problem with the game right now is that nothing makes the player break out of that stalemate. A strategic situation doesnt HAVE to mean HOLY CRAP USE 50 GOD TOKENS. If the player can solve it by strategically building a couple of new cities on the blue side, then hey, that's good too, and even that STILL throws off the balance in the way that should be happening with the game. But nothing right now even pushes it to THAT point. Bandits are a danger, but they're weak by themselves, they need to be able to add to something that's already going on.
2. One problem I have with the idea of Cataclysms really is the extreme randomness. But not for the reasons you might think. My logic is simple: If a Cataclysm *isnt* happening, or one of them DID happen but ended up getting thwarted quickly without doing too much, then chances are..... boredom is happening. At this point, the player would be back to "stalemate is happening, balance not in danger: do nothing but tweak things every now and then and slowly build up both sides equally" until another one came. During this time, they can probably very safely do literally nothing other than build up both sides at the same time (decreasing their chance of being in danger of losing ever further, lowering the threat level of most things). This was one of the reasons why I was so much for an idea like the VPs, which would be there as a constant during each game, as they give me something to CONSTANTLY work towards (preferrably with inherantly inbalanced goals).... and with this game, constantly having to do things means a constant risk of a player action setting stuff off, which to me is the interesting bit with the game.
3. The other problem: The early game is likely to be really dull (as it is right now). I'm not a fan of the Warcraft-esque "The real fun starts when you get to level 23895274589!!!!11" sort of idea. If a game always has a boring stretch at the start where it takes awhile to get to the REAL gameplay (as in, where stuff actually HAPPENS and you have to think).... something is, to me, inherantly wrong with the design. The wait for a Cataclysm or whatever to occur for the first time would be exactly this. Some players can put up with that sort of thing in games.... in my case, it'd be an instant cure for insomnia. Hell, I'd rather start the game IN THE MIDDLE OF a small cataclysm, than have that happen. The idea of "Ok, here's this map, now do your usual basic build-up that you do every game while you wait for the actual challenging bits" isnt interesting at all, but "Ok, here's the map, now do your OH GEEZ HALF OF IT IS ON FIRE", while a little absurd, is at least something HAPPENING.
4. To *really* work, Cataclysms would have to be specifically imbalancing. IE, hurting one side.... but not really hurting the other. Think of a map with the red and blue guys on each end, right? And some horrible boss monster appears exactly in the middle. What is the logical reaction here in a game where maintaining balance is the challenge? Simple: one side does one thing.... and the other mimics it. Squashing the boss/monster/threat/chainsaw skeletons in the middle by smacking together 2 sides that are, despite the monster, of mostly-equal power is.... still boring, and would be an easy strategy. With enough randomness, I dont actually think THIS point should really be a problem, but it popped into my head as a theoretical issue, so I figured I'd mention it anyway. I like being thorough.
5. This one actually strikes me as a big problem: The size of each side. If the player has enough time to build up both sides into being really darn massive.... the threat of truly losing is likely to vanish, because there's simply too much to destroy in the time remaining. It's kinda like what happens in many 4x games; there's alot of strategizing and planning and cool stuff..... up until a certain point, maybe 2/3rds of the way through (or some other arbitrary fraction), and then.... it's a boring steamroller, because you have achieved the point where your own defeat is impossible, yet the game is still going. I love 4x games, but i have a hard time FINISHING a game of one exactly for this reason. There's no strategy or challenge whatsoever when that sort of thing happens. And it's possible in this game, to get BOTH sides SO strong (in an equal fashion) that even firing off a pile of god tokens, while damaging, doesnt put you in danger of losing completely. And with this one, I come to:
6. As I understand it, the losing condition is when a side is *totally* wiped out.... that actually might be a tough condition to hit! In my time with the game, the worst it has ever gotten is me being sorta close MAYBE to almost kinda losing half of a city possibly. And that's WITH me experimenting with the larger powers and throwing things off. This is the OTHER reason why I liked the idea of a different "lose condition", or thing that you had to accomplish to win, because then it's NOT just about "be super efficient and spam 20 zillion units on each side, and then screw the balance, it doesnt matter anymore, you have too many things, it's not possible for anything to kill them fast enough in the time remaining". Heck, even just taking ONE city can take quite alot of units to do in a decent amount of time.
So, that's my thoughts on that. I know I sound negative, but it's my job to be negative. I am "Misery" after all
I chose that name for a whole pile of reasons. If I'm not being at least vaguely negative, I'm likely asleep.
Anyway, I think the idea has potential, but I do think there's some holes in it that need to be worked out first. If done right though, this should be pretty entertaining.