Capitals are usually good, and they can usually pay for themselves, once they grow back to a reasonable population, can't they?
Right. But don't forget - my first loot capital was captured somewhen in 3???bc. I even had no worker when captured it. So markets is of course good option but unfortunately I did not research them yet in 3???bc. Cottage spam + sailing + courts allowed to let the captured cities grow but not before - and markets which came thousands of years later allowed a further continuation of expansion.
Leave one AI some room for development, and then when the time is right...
Did not like this option at all. My cities would be too distributed (hard to defend), and cottage-spammed (good booty if someone decide to pillage them). So, I decide to kill them both. Disadvantage - I was not able to trade techs for a long time. These two poors provided me with what they know but they were not too much succesfull in research anymore. Advantage - I did not need spend tons of shields on military and tons of gold on their upgrade.
bring your trained troops to take over their already developed cities.
I did not allow AI to start the expansion, but barbarians started. There was a great deal of well-developed barbarians cities in 1000ag. So, it makes no difference. Advantages and disadvantages are above. When I play Incas, I always try to kill anobody I meet in 3???bc (while quechas brand is not outdated) and capture their cities and workers. Check how much a settler/worker costs and compare it with cost of quechas. This will demonstrate you that actually such ancient age wars are very profitable affairs from cost & benefits analysis. Now I just improved a general strategy by introducing "great generals" factory in the concept.
Will try it on the emperor, monarch now is too easy.
Isn't one military factory good enough?
right you are! If I repeated the same game, I would not wait for when my first victim would establish a second city. I would capture its capital immediately and will go after the second civ following my plan. Actually one factory is enough. AI starts building archers like crazy using whipping as soon as the 'factory' reaches 2 population and he invents bronze. However, you pillaged him and he never will be able to build axeman or swords. Archers, tens of them, hundreds of them. Before catapults, 2-3 quechas fortified near city in forest and attacking each unit which came out - is enough. After you've catapults, you need 3-4 catapults/trebs and 2-3 quechas/axemans to make the factory work (4-5 points per turn). Maybe one catapults more, so they might heal after attacking of fresh archers without pause.
If you've two factories - it will generate more military points (advantage) but you will start paying a lot of for your military support. A lot means 5-6 of golds. In modern wars, this would be just nothing but in that time, 5-6 of golds means 10% of my total research!
So, I agree - the second factory was unnecessary.