Not really. The "real" empires were the inventions of people in the first place. It is very possible that someone would want to invent the "Empire of Britannia" to further their own prestige.
That's true but invented titles are titular, like the Kingdoms of Trinacria, Cybrus, Nubia and a couple more, and a number of duchies in historical starts. The very idea of de iure is that it is the opposite of inventing something to further your own prestige.
Nothing wrong with the ability to go up from king (even in the form of spawning a titular empire with a custom name and integrating your kingdoms into it de iure after 100 years as per general rules), it's different when you have countries that never existed but pretend they did. If you look at the de iure map in 1066, a supposedly historical start, it shows you the supposedly already existing concept of a legally sound empire of Scandinavia, Francia etc., on par with the real kingdoms of Lotharingia or Burgundy that had gone into desuetude (actually, Burgundy was legally extant just not used by the holder much but anyway).
The fantasy empires are all possible, in theory, even if some of them are extremely unlikely. If history is wild, then wild things have to actually happen.
Again, the bone of contention here is the de iure map. Nothing wrong with there being a mechanism for a king to become an emperor after meeting some plausible conditions. On the other hand, much is wrong when the de iure map is full of fantasy.
I would not call it either balance or candy, but gameplay. The choice between historical accuracy and good gameplay is usually an agonizing one. I respect both approaches, having a great deal of experience looking at game design from either angle. I understand the historical arguments. However, though history might be King, gameplay... is Emperor.
The way I see it, if you meticulously reconstruct the provinces, dig through the lists of rulers of every little piece of land, providing ancestors, predecessors etc., assign the appropriate cultures for the appropriate provinces and game starts, so that one can start from any actual
day of his choosing (like I picked 8 Feb 1296 in my last game), it all seems to be negated by a U-turn when you put fantasy empires on the map. (As opposed to making them creatable via events, plots, decisions etc., even with instant de iure assimilation.)
As for the gameplay value or learning curve of it, one still needs to go to landed_titles.txt to read the conditions, otherwise one can keep looking at the map in frustration, expecting them to work like standard (50% of the counties)... I don't buy it. Other than making an empire in every corner of the map so that all 4-5 human players in a multiplayer game can be emperors, I don't see the benefit, while I see so much which is lost by allowing such a large degree of fantasy in a historical game.
