If your programmers are that good, then use those skills to deliver on time and under-budget. And between now and the start of your next big project, think about how to better leverage those skills your team has, with the bigger budget that comes with a good track record.
But if you must do things this way, then pick ONE cool thing. Not all, not even two -- Never introduce too many risk factors at once. And that one that you pick MUST be somehow central to the gameplay, because all the rest is just fluff. When you're Blizzard, you can afford to sit around adding neat features -- although their decisions are always business-minded IMHO.
But trying to implement all or even just a few of the things your coders can do, because it seems cool and you kinda-sorta think you can, will head you for a very big fall.
Again, the key is DO NOT add anything that will not with certainty contribute to the core game dynamic -- whatever that is (it sounds as if that has yet TBD).