It's not so much about WHAT'S being combined, as HOW.
Even when you're breaking new ground, it's possible to make good decisions about how you're going to approach it. CR managed to orchestrate a perfect storm of fail:
Step 1: Pick an engine that's fundamentally unfit for purpose.
Step 2: Before having the slightest clue what an overarching technology framework might look like for this kind of game world, or even the most rudimentary spec, ramp up production across several studios on gameplay 'modules', final-quality ship models and textures. For the layman, this is roughly equivalent to asking different people in different countries to make different but interlocking parts of a LEGO model without first defining what the model is supposed to look like or indeed what LEGO is.
Step 3: Relentlessly stir and churn the ensuing cacophony with impulsive feature-creep, demands for throw-away demos and obsessive micro-management.