I read fieldfield talking today about having decreasing gathering rates for each additionnal units put in a single feld, was that related ?
(it might be someone else but I read somewhere else that historical accuracy is important only as long as it doesn't have a negative impact on the wordplay).
Introducing negative externalities/decreasing returns could help to reduce the exponential growth effect. For example, we could also have a lower gathering rate for each additional units using the same storehouse. Then how many dropsites you build wouldn't be just a question of distance to ressources but depends on the number of workers using it (dropsite should then also be balanced between civ).
=> More thinking = more fun
The following isn't related directly to the mod, but it might be considered while implementing changes.
I am wondering about how market barter rate are currently determined. They usually seems quite ok in 1vs1 but in tg, they often get crazy/unfair and can easily lead to game imbalances. In some recent games, I have managed to get to barter efficiency close to 300% thanks to Brennus or some additional mines. In the summary attached, I came to choose a strategy in which I would keep the ptolemies enemy player alive as long as possible just to artificially raise metal prices.
The strategy is quite simple:
1 - Send army for direct confrontation with the mineral dependant player, it doesn't really matter if it is a good fight or not as long as the enemy loose a lot of units with a high cost in mineral;
2 - keep sending units until the enemy can't sustain the fight because he runs out of minerals (because he can't gather it fast enough or he is running out of it, in both cases, he would turn to the market anyway)
3 - once he doesn't react too much anymore, just turn to attack some other enemy, wishing his allies will send him food/wood in the meantime that he will trade for minerals in order to rise his pop again. As soon as it has some new mineral costly units, come back to put more pressure on him and on market prices.
With prices on minerals increasing a lot, you can basically keep just a few women on food (to use for protection against sneaky rams too), delete the rest and leave some men on minerals to gather what you will barter later. The larger your army is, the easier it is to put pressure on mineral prices, and the less units you need then on eco since you get better barter rates, the less units you loose also while fighting because you have a larger army, the less ressources you need to make new units...
Civilizations like ptolemies or athens should be very strong in team game, but sadly, they often just end up looking like junkies. Even kushites could potentially become close to fun with enough metal...
Changes in the way map are generated could help, this was already said elsewhere. Another fix (simple but maybe only partial too) would be to changes market barter rates determination. Barter rates of different teams should be independent, it doesn't make too much sense that how much ressources a player is bartering affects the barter rate of his enemy. I don't think they are supposed to barter together.
I also think that market prices volatility is partly responsible for the sustainability of a relatively small number of mineral dependent civs in the same team game. Markets should offer a way for those civ to hedge themself against map-gen risks. Any insurance should have a cost, but if there is a large need for barter and barter efficiency falls too low, then this cost becomes too expensive in the current version of the game (another S curve needed there??) .
Currently, traders seem like an unreliable suppliers to compensate for prices volatility (slow to put in place, hard to protect constantly...). Maybe a few other simple adjustments could be made to improve the game from this perspective, I just gave some ideas above...
I was also wondering about a last thing. Would it be complicated/feasible to introduce different land type on the same map ? I do like that currently wood is slowing down units and is preventing sieges from passing. Having some kind of muddy land type which simply slow down all units could introduce a fun strategical dimension to the game. Scooting would give a higher reward, range units would have a bigger advantage in fighting over this type of land etc...