I don't know the exact formula for return on investment. In the spirit of Civ 5 and not knowing what you are doing I may not want to know. However, as a rule of thumb I expect that as long as the city has more population than tiles it took to road up to them, you are making some minimum profit, which will then grow as the city grows. I, too would have not built so many roads, but it was the team's decision and seems to be working out. That has become a learning experience for me.
You're almost right: the formula for trade route profit is (pop * 1.25) + 1, so you actually need one less pop than road tiles required to be profitable. All the routes from Rome to our closest cities are profitable at this point, and will only become better with time.
I think the question should be WHEN to build roads, not IF. In general, its probably a good idea to improve resources around new cities first, and then hook them up once they've grown a bit. EDIT: ohh, and I think we can still improve the road layout to save some maintenance.