Ok, a detailed explanation of why the “city momentum” factor for Kohan doesn’t exist. Counterattacking to get cities back is really, really easy, the advantage you get in short-term production for a city is amazingly minimal, and you normally have so many cities it doesnt matter.
Things you get for taking the city:
Against a decent human player, you’re lucky to ever capture a city with a
single building intact. It’s quite obvious when you’re not going to be able to hold the city, and selling all the improvements takes about 5 seconds. So, no buildings.
Unless your opponent is following up with a fully trained team of engineers, cities tend to be a state of disrepair (no hp, no militia) already for a counterattack, lowering the bar for capture.
You get 5 or 10 gold a minute once you capture the city, and production from anything in the support radius, but this is usually really minimal.
You do, however, get access to any extra building slots your opponent has added to the city. This isn’t that big of an advantage, though, as it takes a pretty long time to fill up the slots, and even longer for them to pay for themselves. Also, you need to be absolutely certain that you can hold the new acquisition if you want to start filling it up with buildings; selling off stuff you’ve held for a long time isn’t that painful, but man, does it suck to build a bunch of stuff only to sell it at a 50% loss three minutes later.
You get a few troop slots, but this is also meaningful in the first five minutes of the game or so.
Things your opponent loses:
The income from the city’s buildings, in resources or gold. However, they get 50% or so cost back for selling them, so this isn’t that big of a hit either, especially when you consider they’ve been getting production out of them for quite a while.
The income from the city itself. See above, this isn’t worth anything.
The troop support space. See previous.
So, in summary, when you capture a city you get an empty, undefended, burning building, without any nearby defending forts that were nearby, that would have cost you, oh, something like 150 gold to build yourself, assuming a three-slot city, which most are.
Losing a fully upgraded mega-city will cripple you, yes, but most players have one or two of those, they’re at the center of their empire, and they’re impossible to take unless you’re about to win the game anyway. Hell, in the multiplayer games I’ve played, I probably build, on average, 20 cities a game. This isn’t like Warcraft where you tend to top out at 4 or so cities or gold mines a game; it’s far too easy to just spread like the clap.
If I remember correctly, upgrading to a level 3 city is only “worth it,” in terms of investment, for Council. The other sides get better bang for the buck by never going beyond level 2 unless they need to make a troop factory town.