Suggestion to fix the conflict between champions and trained units.

Hi! I'm a long time lurker on these forums and I've been playing this game on and off since WoM. The game is great at the moment and really seems to be shaping up to be as good as Galciv (which in my opinion is a fantastic strategy game). However, I do have a few suggestions to solve some of the lingering problems which still exist with regard to champions, trained units and lack of hard choices or reward for specialisation at the grand strategy level. I realise it is a little late in the process to be suggesting changes and some of these suggested changes will sound quite major however they should be easy enough to implement (I think) and should complement the focus and feel of the game while solving several problems simultaneously.

I have a nasty flu at the moment so please forgive any mistakes or lack of clarity in my suggestions! Also apologies in advance for the wall of text below!

The problem

Balancing the power level of champions versus trained units is extremely difficult under the current system as there is almost no interconnection between the champion/adventuring side of the game and the city building/ unit training side of the game. An additional problem that results from this is lack of differentiation from game to game at the grand strategy level. The same strategy of leveling your champions by attacking monsters and spamming settlers followed by leveling your cities always works. There is no need to scout your opponents as you will almost always face the same armies of crappy trained troops accompanied by samey heroes with mid-level magic and mid-level combat abilities regardless of which paths they have chosen during leveling up.

The solution

The conflict between champions and trained units can be resolved by linking the recruited champions to production and allowed army and unit size in a mutually exclusive fashion. This forces the player to make a hard choice with a strong opportunity cost each time they recruit a champion. This incentivises players to decide if they want to have a strong focus on super-champions, armies, or production/research and rewards them for specialising and thinking carefully about which path they choose for their champions

Specific suggestions

Army size is removed from the tech tree and instead becomes part of the trait system under a new path; Path of the General. Path of the general replaces path of the defender. Unit size is also removed from the tech tree and becomes a trait under a new path ; Path of the Administrator, which replaces path of the Governor.

Path of the General

Maximum size of an army with no hero attached reduced to three units. The max size of an army with any hero attached becomes four (i.e. the hero + three units). Only heroes with path of the general get traits which allow army sizes larger than four units (e.g. a general with the new  trait Logistics I can have five units in his stack, ect.). Path of the general replaces path of the defender. Current path of the defender traits moved into path of the warrior. All traits which improve the army as a whole, e.g. all units in the army get +10% accuracy, are moved into path of the general. This means that players who want large, powerful armies will have to specialise heroes as generals, at the expense of that hero's combat abilities while players who want powerful Sauron-like heroes will have small armies.

Path of the Administrator

Current path of the governor is replaced with path of the administrator.  Unit size techs are removed from the tech tree. Heroes with path of the administrator gain traits which allow the max unit size trainable in that city to be increased (e.g. only three in a unit in a normal city while a city with a hero with the new drillmaster I trait stationed can train units with five members ect.). Other administrator traits should  give very significant bonuses to research (loremaster) , money (merchant) and production (overseer) in the city that unit is stationed in. These bonuses should be comparable to, or even exceed the bonuses from enchantments so that players who want lots of gilder, high production cities or stronger armies have a large incentive to park heroes and let them level inside the city. The tech tree should gain buildings which provide significant experience gain on a per turn basis to administrators but perhaps not to other paths (though cross-overs such as the war college providing experience to stationed generals and administrators or the library providing experience to stationed  mages as well as administrators may be viable). These improvements should be spread through the three branches of the tech tree to prevent players racing to them too quickly. Ultimately the stationed administrators should gain experience at a similar rate to what the other champions gain in the field.

Advantages of this system

The main strength of this system is that it is full of  opportunity cost. With this system when you get a new hero you can decide that rather than having a combat hero you can use him to improve your civilisation. By making appropriate trait choices you can use him to improve your army quality (via unit size and superior research), your research, production or economy (all being viable choices towards victory) by making him an administrator and parking him in a city. If you choose to have a combat hero you can now choose whether you want one who is extremely strong by himself (warrior, assassin or mage) or one who is relatively weak but can command large armies and provide strong bonuses to them.

You can imagine a civilisation which has focussed exclusively on combat heroes having a few champions wandering around with demi-god-like combat abilities and magic but relatively poor cities and small, crappy armies. You can contrast this to a civilisation with lots of generals who are individually weak but are commanding large, highly buffed armies, versus one who has concentrated on administrators and whose armies are small but could be numerous (due to production bonus), well  equipped (due to superior research), or tougher (due to larger unit size). Obviously mixed strategies are also very viable and hopefully the best strategy at any one time in a particular game will require some thought.  This increases the number of meaningful choices available to the player and allows heroes to be used to further strategies other than simply wandering about bashing everything they encounter.  By tying army size and quality to the hero trait system it should be significantly easier to balance armies versus heroes without nerfing one or the other into uselessness, than the current system where heroes exist essentially independently of the cities and armies of the civilisation that they belong to.   

This system also allows players to compensate to some extent for poor starting positions i.e. ones with low production, by stationing an administrator and choosing overseer traits (+percentage production per materials). This may help solve the problem of players valuing starting positions with high materials more than those with high grain.

Most of the changes mentioned should be relatively easy to implement with only changes to the xml required (I think?).   

Disadvantages

Champions will need to be rebalanced so that recruiting a late game champion doesn't suddenly give an enormous bonus for very little investment. Increasing their gilder cost substantially and adding an influence cost might be a  good idea. My preferred system would be three options given when you try to hire a champion; an extortionate gilder cost that only an empire with a strong economic focus can afford, a slightly smaller influence cost, or a difficult quest which only combat-focussed champions or large armies could do, whose reward would be getting the champion for free; "You want me to work for you? Kill Merduul who I hate and I'll be happy to join you...". This system would probably be a lot of work though so perhaps providing more gildar as quest rewards (in addition to items) would prevent combat focussed empires from falling behind in ability to hire champions.

Low level champions whose paths can be chosen after hiring will also need to be more plentiful to ensure that what should be a strategic choice (i.e. which paths to choose for your champions) doesn't become pot luck