What follows is my take on Fallen Enchantress' balance and its failings. I admit it's completely subjective, and that I might not understand the ideas behind the current balance choices well enough. But after playing so many games, this is the conclusions I came to and the feelings I was left with.
Fallen Enchantress faces two strong issues, mainly due to its numerous components. Those issues are obsolescence (that is, the tendency of content and components of the game to lose their usefulness at a sustained pace) and relatively poor scaling/control of the pace.
These issues are linked, but aren't identical. Obsolescence can mainly be felt in terms of armies and monsters. When military technologies are discovered, they invariably render all former military techs completely obsolete. There is no point using weapons from previous technologies anymore, or smaller groups - in the end, the solution is always to put the best weapon and the best armor on the largest group possible. There were a lot of dev' diaries about "quantity vs. quality" and how these would be balance - this is just not the case. For armies, quality is invariably better, because the increase of cost for better quality soldiers is much less than the increase in power and survivability.
Civilization, another long game with many more military technologies than Elemental, also features obsolescence. But Civilization doesn't feature customizable units, lasts a very large number of turns (much larger than the time it takes to reach chain/plate armor, when an FE game is usually over), and obsolescence comes much more gradually with less emphasis on quality (some, but less).
This also creates a problem with recruited and special units, such as Darklings and Resoln's spiders. Their window of usefulness is short, and if you're unlucky, they'll be obsolete before you can actually find a recruitable "resource". Again, this leads to content becoming poorer.
In the end, I believe the game would be much more interesting if all of these choices stayed relevant for a much longer time (for the whole game, maybe? At least a bigger part of it), and if quality didn't dominate the game so much in terms of military power. And it would be much more interesting if all those choices had unique characteristics to them, a role that they would fulfill better than the other options. And the AI would fare much better if its equipment could compete with the player's, which it almost never does - because if it's a bit late on the player technologically, its armies will be slaughtered due to the huge jump in power (for a minimal increase in cost) this translates into.
The scaling problem comes from the lack of regulating mechanisms in the game. It's hard to know how many shards or cities a players will have - so how does one balance magic or research? Civilization IV had maintenance, Civ V has tech/policies prices increase and happiness costs (that can be overcome but only after a while). Sure, these aren't perfect, but they do help reduce issues by keeping research and resource production within a certain range. FE has almost none of this - you can have many cities and they can all be big, and it's hard to know what a typical player's research output will be. Similarly, it's hard to know how many shards (or iron, or any other resources) that player will have. As a result, I find the game to be really unbalanced and difficult to balance.
For instance, how does one balance a fire damage spell in the current paradigm? Too much importance on the shards, and if they get 5 of them they will slaughter everything in sight. You can reduce base damage to compensate, but if the player can't find fire shards (happens very often, due to the high variability in starting conditions), the spells will then be useless - and if his sovereign is a fire mage, this severely weakens him. Too little emphasis on the shards and you break their usefulness. In the end, because there is no control over the variability in resources the player will have access too, because of the very unstable scaling in Fallen Enchantress, balance has become a small nightmare.
The same happens to the difficulty of monsters: it's very hard to make monsters that will be adequately challenging because it's hard to predict what sort of power the player will have access to during a typical encounter. Of course, some variability *is* good, but in Fallen Enchantress, it's completely out of control. Hence these regular, contradicting complaints that things are either too hard or too easy... for the same patch. When you know how to harness the game mechanics, the game becomes easy and a few elements stand out as much more powerful than the others - you can exploit the lack of reasonable scaling. Otherwise, the game is much harder.
The absence of a limiting mechanism on the number/size of cities (big cities, or lots of cities? In FE, you can - and should- have both) makes the regulation of research difficult. If players expand a lot, they will have a lot of research and will be able to produce better troops - making the AI even less able to put up a fight.
It might be outside of Beta 5's scope to introduce regulating mechanisms and consider how to keep each side of the game within reasonable limits. Most of these issues can be modded away, with work and ingenuity - however, I am afraid the AI will not be able to play the game well if these changes are too radical - and that would make the game less interesting, not more.
